Breast cancer is a common malignant tumour and studies have shown that early and accurate detection is crucial for patients. With the maturity of medical imaging and deep learning development, significant progress has been made in breast cancer classification, which greatly improves the accuracy and efficiency of classification. This review focuses on deep learning, migration learning, GAN, and lifelong learning to elaborate and summarise the important roles arising from breast cancer detection. This review also examines the dataset and labeling issues required for breast cancer classification. In conclusion, at the end of the article, we look at future directions for breast cancer classification research, including cross-migration learning, multimodal data fusion, model interpretability, and lifelong learning, and also explore how to provide personalized treatment plans for patients.

At the beginning of 2020, with the onset of the pandemic, the traditional learning environment for learners drastically changed globally. Since then, most students/teachers have started and practiced distance and virtual learning/teaching. Thus, a technological breakthrough in virtual learning has followed. In connection with this, many countries worldwide have commenced allocating additional financing and funds for educational institutions' technological improvement and development. The long-term stay in distance learning has revealed and highlighted new problems students face: their knowledge level has decreased, they lack socialization skills, and they face psychological and physical health problems. Due to this negative impact on students, a need to research and evaluate how much the EU countries allocated to solve the distance learning-caused problems and what programs or models they prepared has emerged and encouraged further studies. The research has found that many countries increased their allocations very minimally, e.g., 0.0.1%, but some increased their available budgets to 32%. Notably, most countries did not separate distance learning exit funding from distance learning preparation funding. Based on the problems the countries saw, only a few states identified withdrawal from distance learning as a problem. Considering this, we set ourselves the goal to evaluate exit models from distance learning and allocated funding amounts. The following objectives were planned to achieve the goal: · to evaluate the global practice of exit from distance learning; · to determine the scope of funding for pandemic management; · to evaluate the amounts of funding allocated to manage pandemic-caused consequences and the GDP ratio. Research methods: mathematical-statistical analysis, empirical analysis, and analysis of scientific literature.

Reinforcement Learning (RL) has demonstrated considerable potential in solving challenges across various domains, notably in autonomous driving. Nevertheless, implementing RL in autonomous driving comes with its own set of difficulties, such as the overestimation phenomenon, extensive learning time, and sparse reward problems. Although solutions like Hindsight Experience Replay (HER) have been proposed to alleviate these issues, the direct utilization of RL in autonomous vehicles remains constrained due to the intricate fusion of information and the possibility of system failures during the learning process. In this paper, we present a novel RL-based autonomous driving system technology that combines Obstacle Dependent Gaussian (ODG) RL, Soft Actor-Critic (SAC), and meta-learning algorithms. Our approach addresses key issues in RL, including the overestimation phenomenon and sparse reward problems, by incorporating prior knowledge derived from the ODG algorithm. We evaluated our proposed algorithm on official F1 circuits, using high-fidelity racing simulations with complex dynamics. The results demonstrate exceptional performance, with our method achieving up to 89% faster learning speed compared to existing algorithms in these environments.

This review paper delves into using micro-credentials in higher education ecosystems as a digital enablers. Micro-credentials, which are digital credentials that attest to a learner’s mastery of a specific skill or knowledge area, are becoming more popular in higher education. The paper examines the successful implementation of micro-credential frameworks in higher education, using case studies to demonstrate the advantages of micro-credentials. The review emphasizes the agility and flexibility of microcredentials, which enable learners to acquire new skills quickly and respond to changes in the job market. In addition, the paper discusses the digital nature of micro-credentials and how they allow institutionsto provide targeted, skills-based training that isrelevant to employers. It also explores how micro-credentials are delivered through online platforms, making them convenient and easily accessible for learners. The review underscores the significance of digital infrastructure, connectivity, and public utility for promoting micro-credentials. The paper argues that micro-credentials function as a digital enabler for higher edu- cation ecosystems, allowing learners to acquire targeted training and enabling institutions to expand their offerings and reach more students. The paper concludes by highlighting the potential for micro-credentials to help bridge the skills gap and equip learners with the skills necessary to succeed in today’s digital economy.

Tuberculosis (TB) disease still remain a major global threat due to the growing number of drug-resistant species and global warming. Despite the fact that there are new molecular diagnostic approaches, however, majority of developing countries and remote clinics depends on conventional approaches such as Tuberculin test, microscopic examinations and radiographic imaging (Chest X-ray). These techniques are hindered by several challenges which can lead to miss-diagnosis especially when interpreting large number of sample cases. Thus, in order to reduce workload and prevent miss-diagnosis, scientists incorporated computer-aided technology for detection of medical images known as Computer aided Detection (CADe) or Diagnosis (CADx). The use of AI-powered techniques has shown to improve accuracy, sensitivity, specificity. In this review, we discussed about the epidemiology, pathology, diagnosis and treatment of tuberculosis. The review also provides background information on Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), Transfer Learning (TL) and their applications in detection of tuberculosis from both microscopic slide images and X-ray images. The review also proposed an IoT/AI powered system which allows transfer of results obtained from DL models with end users through internet networks. The concept of futuristic diagnosis, limitations of current techniques and open research issues are also discussed.

Intelligent algorithms are recently used in the optimization process in chemical engineering and application of multiphase flows such as bubbling flow. This overview of modeling can be a great replacement with complex numerical methods or very time-consuming and disruptive measurement experimental process. In this study, we develop the adaptive network-based fuzzy inference system (ANFIS) method for mapping inputs and outputs together and understand the behavior of the fluid flow from other output parameters of the bubble column reactor. Neural cells can fully learn the process in their memory and after the training stage, the fuzzy structure predicts the multiphase flow data. Four inputs such as x coordinate, y coordinate, z coordinate, and air superficial velocity and one output such as pressure gradient are considered in the learning process of the ANFIS method. During the learning process, the different number of the membership function, type of membership functions and the number of inputs are examined to achieve the intelligent algorithm with high accuracy. The results show that as the number of inputs increases the accuracy of the ANFIS method rises up to almost for all cases, while the increment in the number of rules has a effect on the intelligence of artificial algorithm. This finding shows that the density of neural objects or higher input parameters enables the moded for better understanding. We also proposed a new evaluation of data in the bubble column reactor by mapping inputs and outputs and shuffle all parameters together to understand the behaviour of the multiphase flow as a function of either inputs or outputs. This new process of mapping inputs and outputs data provides a framework to fully understand the flow in the fluid domain in a short time of fuzzy structure calculation.

Undernutrition and micronutrient deficiencies such as anemia are considered significant public health challenges in Bangladesh, which enhancing fish consumption is a well-established food-based intervention to address these. This paper documents the establishment of community-based fish chutney production and reports the impact of its consumption on mid-upper arm circumference (MUAC) and hemoglobin (Hb) levels among targeted 150 pregnant and lactating women (PLW) in rural Bangladesh. A fish chutney was developed using locally available ingredients followed by a series of laboratory tests, including nutrient composition, shelf-life and food safety. A community-based fish chutney production process was designed to: 1) supply locally available ingredients for processing; 2) establish two fish drying sites; 3) initiate a community-based production site, and 4) distribute fish chutney to PLW by six women nutrition field facilitators for one year. Then a pre- and post-intervention study was designed for a selected 150 PLW to receive 30g of fish chutney daily for 12 months. Differences in mean mid-upper arm circumference (MUAC) and hemoglobin (Hb) levels pre- and post-consumption were analyzed using one-way analysis of variance. Consumption of 30g of fish-chutney resulted in significant increases of the mean values of Hb levels and MUAC among the targeted PLW.

Video self-supervised learning (VideoSSL) offers significant potential for reducing annotation costs and enhancing a wide range of downstream tasks in video understanding. The ultimate goal of VideoSSL is to achieve human-level video intelligence across a spectrum of tasks, from low-level tasks such as pixel temporal correspondence to high-level complex spatio-temporal tasks like action recognition. However, most existing VideoSSL methods focus on isolated aspects of this spectrum and fail to integrate different levels of task complexity. Our study presents the first comprehensive survey that connects all families of VideoSSL methods. We provide a detailed review of the full spectrum of VideoSSL, from low to high levels, by conceptually linking their self-supervised learning objectives and including a comprehensive categorization. Our extensive evaluation highlights the strengths and limitations of each SSL objective across various downstream task families. We also detail the challenges in current VideoSSL research such as data curation, interpretability, deployment, and privacy concerns, an area that previous surveys have not thoroughly explored. In addressing these challenges, we recognize the strengths of existing methods in addressing these challenges and outline future directions for research.

Recent advancements in artificial intelligence (AI) and deep learning have revolutionized the field of protein engineering, particularly in the area of de novo protein design. This review article explores the impact of AI-driven approaches on protein design, with a specific focus on the role of structure prediction neural networks, such as AlphaFold and RoseTTAFold. The article discusses the paradigm shift brought about by these networks, which have enabled the design of proteins with unique structures and functions that are not found in nature. The review covers various aspects of AI-driven protein design, including the use of protein language models to harness evolutionary information, the development of de novo protein design workflows utilizing deep learning, and the application of generative models like RFdiffusion and RoseTTAFold All-Atom. The article also highlights the successes and applications of AI-driven protein design across diverse domains, such as enzyme engineering, antibody design, and vaccine development. Additionally, the review identifies current challenges and future directions in the field, emphasizing the need to address limitations in modeling conformational dynamics and designing proteins for in vivo functionality. The article concludes by underscoring the potential of AI-driven protein design to transform various aspects of science and technology, while also acknowledging the importance of interdisciplinary collaboration and the development of robust pipelines for the validation and optimization of designed proteins. Overall, this comprehensive review serves as a valuable resource for researchers and practitioners interested in understanding the current state and future prospects of AI-driven protein design.

Artificial intelligence (AI) refers to computer algorithms that replicate the cognitive function of humans. Machine learning is widely applicable using structured and unstructured data, while deep learning is derived from the neural networks of the human brain that process and interpret information. During the last decades, AI has been introduced in several aspects of healthcare. In this review, we aim to present the current application of AI in the neonatal intensive care unit. AI-based models have been applied to neurocritical care, including automated seizure detection algorithms and electroencephalogram-based hypoxic-ischemic encephalopathy severity grading systems. Moreover, AI models evaluating magnetic resonance imaging contributed to the progress of the evaluation of the neonatal developing brain and the understanding of how prenatal events affect both structural and functional network topologies. Furthermore, AI algorithms have been applied to predict the development of bronchopulmonary dysplasia and assess the extubation readiness of preterm neonates. Automated models have been also used for the detection of retinopathy of prematurity and the need for treatment. Among others, AI algorithms have been utilized for the detection of sepsis, the need for patent ductus arteriosus treatmnet, the evaluation of jaundice, and the detection of gastrointestinal morbidities. Finally, AI prediction models have been constructed for the evaluation of the neurodevelopmental outcome and the overall mortality of neonates. Although the application of AI in neonatology is encouraging, further research in AI models is warranted in the future including retraining clinical trials, validating the outcomes, and addressing serious ethics issues.

Distance Learning has become the new standard, especially after the pandemic and due to the technological advances, that are incorporated into the teaching procedure. At the same time, the augmented use of the internet has blurred the borders between distance and conventional learning. Students interact mainly through LMSs, leaving their digital traces that can be leveraged to improve the educational process. This work aims to propose a model that can capture the students' behaviors based on the clickstream data associated with the discussion forum and additionally to suggest interpretable patterns that will support education administrators and tutors in the decision-making process. To achieve our goal, we use Social Network Analysis (SNA) as networks represent complex interactions in a meaningful and easily interpretable way. Moreover, simple or complex network metrics are becoming available to provide valuable insights into the students’ social interaction. This study concludes that by leveraging the imprint of these actions in an LMS and using metrics of SNA, differences can be spotted in the communicational patterns that go beyond simple participation recording. Although HITS and PageRank algorithms were created with completely different targeting, it is shown that they can also reveal methodological features in students’ communicational approach.

Artificial Intelligence (AI) is becoming increasingly important, especially in the medical field. While AI has been used in medicine for some time, its growth in the last decade has been remarkable. Specifically, Machine Learning (ML) and Deep Learning (DL) techniques in medicine have been increasingly adopted thanks to the growing abundance of health-related data, improved suitability of such techniques for managing large data-sets, and more computational power. The Italian scientific community has been instrumental in advancing this research area. This article aims to conduct a comprehensive investigation of the ML and DL methodologies and applications used in medicine by the Italian research community in the last five years.

Documenting cultural heritage by using artificial intelligence (AI) is crucial for preserving the memory of the past and a key point for future knowledge. However, modern AI technologies make use of statistical learners that lead to self-empiricist logic, which, unlike human minds, use learned non-symbolic representations. Nevertheless, it seems that it is not the right way to progress in AI. If we want to rely on AI for these tasks, it is essential to understand what lies behind these models. Among the ways to discover AI there are the senses and the intellect. We could consider AI as an intelligence. Intelligence has an essence, but we do not know whether it can be considered “something” or “someone”. Important issues in the analysis of AI concern the structure of symbols -operations with which the intellectual solution is carried out- and the search for strategic reference points, aspiring to create models with human-like intelligence. For many years, humans, seeing language as innate, have carried out symbolic theories. Everything seems to have skipped with the advent of Machine Learning. In this paper, after a long analysis of history, the rule-based and the learning-based vision, we propose KERMIT as a unit of investigation for a possible meeting point between the different learning theories. Finally, we propose a new vision of knowledge in AI models based on a combination of rules, learning and human knowledge.

Using data-driven models to solve predictive maintenance problems has been prevalent for original equipment manufacturers (OEMs). However, such models fail to solve two tasks that OEMs are interested in: (1) Making the well-built failure prediction models working on existing scenarios (vehicles, working conditions) adaptive to target scenarios. (2) Finding out the failure causes, furthermore, determining whether a model generates failure predictions based on reasonable causes. This paper investigates a comprehensive architecture towards making the predictive maintenance system adaptive and interpretable by proposing (1) an ensemble model dealing with time-series data consisting of a long short-term memory (LSTM) neural network and Gaussian threshold to achieve failure prediction one week in advance and (2) an online transfer learning algorithm and a meta learning algorithm, which render existing models adaptive to new vehicles with limited data volumes. (3) Furthermore, the Local Interpretable Model-agnostic Explanations (LIME) interpretation tool and super-feature methods are applied to interpret individual and general failure causes. Vehicle data from Isuzu Motors, Ltd., are adopted to validate our method, which include time-series data and histogram data. The proposed ensemble model yields predictions with 100% accuracy for our test data on engine stalling problem and is more rapidly adaptive to new vehicles with smaller error following application of either online transfer learning or the meta learning method. The interpretation methods help elucidate the global and individual failure causes, confirming the model credibility.

We discuss the implementation of a suite of virtual diagnostics at the FACET-II facility currently under commissioning at SLAC National Accelerator Laboratory. The diagnostics will be used for prediction of the longitudinal phase space along the linac, spectral reconstruction of the bunch profile and non-destructive inference of transverse beam quality (emittance) using edge radiation at the injector dogleg and bunch compressor locations. These measurements will be folded in to adaptive feedbacks and ML-based reinforcement learning controls to improve the stability and optimize the performance of the machine for different experimental configurations. In this paper we describe each of these diagnostics with expected measurement results based on simulation data and discuss progress towards implementation in regular operations.

The importance of energy systems and its role in economics and politics is not hidden for anyone. This issue is not only important for the advanced industrialized countries, which are major energy consumers, but is also important for oil-rich countries. In addition to the nature of these fuels which contains polluting substances, the issue of their ending up has aggravated the growing concern. Biofuels can be used in different fields for energy production like electricity production, power production or for transportation. Various scenarios have been written about the estimated biofuels from different sources in the future energy system. The availability of biofuels for the electricity market, heating and liquid fuels is very important. Accordingly, the need for handling, modelling, decision making and future forecasting for biofuels can be one of the main challenges for scientists. Recently, machine learning and deep learning techniques have been popular in modeling, optimizing and handling the biodiesel production, consumption and its environmental impacts. The main aim of this study is to evaluate the ML and DL techniques developed for handling biofuels production, consumption and environmental impacts, both for modeling and optimization purposes. This will help for sustainable biofuel production for the future generations.

While the use of crime data has been widely advocated in the literature, its availability is often limited to large urban cities and isolated databases tend not to allow for spatial comparisons. This paper presents an efficient machine learning framework capable of predicting spatial crime occurrences, without using past crime as a predictor, and at a relatively high resolution: the U.S. Census Block Group level. The proposed framework is based on an in-depth multidisciplinary literature review allowing the selection of 188 best-fit crime predictors from socio-economic, demographic, spatial, and environmental data. Such data are published periodically for the entire United States. The selection of the appropriate predictive model was made through a comparative study of different machine learning families of algorithms, including generalized linear models, deep learning, and ensemble learning. The gradient boosting model was found to yield the most accurate predictions for violent crimes, property crimes, motor vehicle thefts, vandalism, and the total count of crimes. Extensive experiments on real-world datasets of crimes reported in 11 U.S. cities demonstrated that the proposed framework achieves an accuracy of 73 and 77% when predicting property crimes and violent crimes, respectively.

The prevalence of differential equations as a mathematical technique has refined the fields of control theory and constrained optimization due to the newfound ability to accurately model chaotic, unbalanced systems. However, in recent research, systems are increasingly more nonlinear and difficult to model using Differential Equations only. Thus, a newer technique is to use policy iteration and Reinforcement Learning, techniques that center around an action and reward sequence for a controller. Reinforcement Learning (RL) can be applied to control theory problems since a system can robustly apply RL in a dynamic environment such as the cartpole system (an inverted pendulum). This solution successfully avoids use of PID or other dynamics optimization systems, in favor of a more robust, reward-based control mechanism. This paper applies RL and Q-Learning to the classic cartpole problem, while also discussing the mathematical background and differential equations which are used to model the aforementioned system.

The world has been evolving with new technologies and advances day-by-day. With the advent of various learning technologies in every field, the research community is able to provide solution in every aspect of life with the applications of Artificial Intelligence, Machine Learning, Deep Learning, Computer Vision, etc. However, with such high achievements, it is found to lag behind the ability to provide explanation against its prediction. The current situation is such that these modern technologies are able to predict and decide upon various cases more accurately and speedily than a human, but failed to provide an answer when the question of why to trust its prediction is put forward. In order to attain a deeper understanding into this rising trend, we explore a very recent and talked-about novel contribution which provides rich insight on a prediction being made -- ``Explainability.'' The main premise of this survey is to provide an overview for researches explored in the domain and obtain an idea of the current scenario along with the advancements published to-date in this field. This survey is intended to provide a comprehensive background of the broad spectrum of Explainability.

Machine learning (ML) has emerged as a critical tool for making sense of the growing amount of genetic and genomic data available because of its ability to find complex patterns in high dimensional and heterogeneous data. While the complexity of ML models is what makes them powerful, it also makes them difficult to interpret. Fortunately, recent efforts to develop approaches that make the inner workings of ML models understandable to humans have improved our ability to make novel biological insights using ML. Here we discuss the importance of interpretable ML, different strategies for interpreting ML models, and examples of how these strategies have been applied. Finally, we identify challenges and promising future directions for interpretable ML in genetics and genomics.

The increasing trend of studying the innate softness of robotic structures and amalgamating it with the benefits of the extensive developments in the field of embodied intelligence has led to sprouting of a relatively new yet extremely rewarding sphere of technology. The fusion of current deep reinforcement algorithms with physical advantages of a soft bio-inspired structure certainly directs us to a fruitful prospect of designing completely self-sufficient agents that are capable of learning from observations collected from their environment to achieve a task they have been assigned. For soft robotics structure possessing countless degrees of freedom, it is often not easy (something not even possible) to formulate mathematical constraints necessary for training a deep reinforcement learning (DRL) agent for the task in hand, hence, we resolve to imitation learning techniques due to ease of manually performing such tasks like manipulation that could be comfortably mimicked by our agent. Deploying current imitation learning algorithms on soft robotic systems have been observed to provide satisfactory results but there are still challenges in doing so. This review article thus posits an overview of various such algorithms along with instances of them being applied to real world scenarios and yielding state-of-the-art results followed by brief descriptions on various pristine branches of DRL research that may be centers of future research in this field of interest.

Crowdsourcing is one of the new themes that has appeared in the last decade. Considering its potential, more and more organisations reach for it. It is perceived as an innovative method that can be used for problem solving, improving business processes, creating open innovations, building a competitive advantage, and increasing transparency and openness of the organisation. Crowdsourcing is also conceptualised as a source of a knowledge-based organisation. The importance of crowdsourcing for organisational learning is seen as one of the key themes in the latest literature in the field of crowdsourcing. Since 2008, there has been an increase in the interest of public organisations in crowdsourcing and including it in their activities. This article is a response to the recommendations in the subject literature, which states that crowdsourcing in public organisations is a new and exciting research area. The aim of the article is to present a new paradigm that combines crowdsourcing levels with the levels of learning. The research methodology is based on an analysis of the subject literature and exemplifications of organisations which introduce crowdsourcing. This article presents a cross-sectional study of four Polish municipal offices that use four types of crowdsourcing, according to the division by J. Howe: collective intelligence, crowd creation, crowd voting, and crowdfunding. Semi-structured interviews were conducted with the management personnel of those municipal offices. The research results show that knowledge acquired from the virtual communities allows the public organisation to anticipate changes, expectations, and needs of citizens and to adapt to them. It can therefore be considered that crowdsourcing is a new and rapidly developing organisational learning paradigm.

Continual learning (CL), a.k.a lifelong learning, is an emerging research topic that has been attracting increasing interest in the field of machine learning. With human activity recognition (HAR) playing a key role in enabling numerous real-world applications, an essential step towards the long-term deployment of such systems is to extend the activity model to dynamically adapt to changes in people’s everyday behavior. Current research in CL applied to HAR domain is still under-explored with researchers exploring existing methods developed for computer vision in HAR. Moreover, analysis has so far focused on task-incremental or class-incremental learning paradigms where task boundaries are known. This impedes the applicability of such methods for real-world systems. To push this field forward, we build on recent advances in the area of continual learning and design a lifelong adaptive learning framework using Prototypical Networks, LAPNet-HAR, that processes sensor-based data streams in a task-free data-incremental fashion and mitigates catastrophic forgetting using experience replay and continual prototype adaptation. Online learning is further facilitated using contrastive loss to enforce inter-class separation. LAPNet-HAR is evaluated on 5 publicly available activity datasets in terms of its ability to acquire new information while preserving previous knowledge. Our extensive empirical results demonstrate the effectiveness of LAPNet-HAR in task-free CL and uncover useful insights for future challenges.

This paper presents an approach to the application of the Model-Based Systems Engineering (MBSE) and Model-Based Systems Architecting (MBSA) principles to develop a Model-Based Pattern Language (MBPL). It takes too long for systems engineers and architects to develop a new system from scratch, particularly new space-based systems derived from the existing space systems architectures. A pattern language is a holistic view of reusable logical model artifacts; many are interdisciplinary and introductory, if at all. The results are mostly a combination of the application-specific logical solution, which further results in the best possible overall solution. The main benefit of the pattern language is reducing the time and validation required to generate a new space-based system architecture; this approach will develop top-level requirements in the initial phase of the system development. The rationale of the methodology proposed by the paper is as follows, collect, and decompose published literature and other open-source information available on space system architectures and system models; develop SysML models for systems, subsystems, products, assembly, subassembly level, and mission-specific requirements using CAMEO SysML software. Arrange these patterns to develop a functional ontology and construct a logical architecture pattern library. This approach created, updated, and managed SysML pattern language, which evaluated the expedited new model construction. Again, our objective is to develop a logical pattern language using public domain information and evaluate patterns by constructing a new space mission concept—for example, planetary surface habitat.

This study introduces the Dynamic Feedback-Driven Learning Optimization Framework (DFDLOF), an innovative machine learning-based approach to personalize educational pathways. DFDLOF leverages advanced algorithms for in-depth analysis of learner data, enabling the dynamic customization of educational content and methodologies. Its application in diverse online platforms, including Khan Academy and Coursera, underscores its adaptability and scalability. DFDLOF enhances learner engagement and outcomes by providing tailored learning experiences and real-time feedback. Key insights reveal its strengths in personalized education, particularly in dynamically adapting learning paths and the importance of real-time feedback. Challenges identified include integration complexities and data privacy concerns. Future research directions emphasize the need for longitudinal impact studies and broader demographic applications. This study contributes significantly to the evolving domain of adaptive learning technologies, demonstrating the transformative potential of machine learning in personalizing education.

Machine learning is a fundamental aspect of artificial intelligence that involves the development of algorithms and models that allow computers to learn and make predictions or decisions without explicit programming. With the development of neural networks, back-propagation algorithms and deep learning, machine learning has made breakthroughs in the fields of image recognition, natural language processing and handwriting recognition using machine learning techniques. The advent of deep learning has revolutionized the field of handwriting recognition, using convolutional neural networks, recurrent neural networks, and sequence-to-sequence models to provide solutions that go beyond machine learning methods and significantly improve the accuracy and robustness of handwriting recognition systems. But challenges remain, including the need for large labelled datasets, computational resources and addressing potential biases. As research in deep learning techniques continues to drive handwriting recognition closer towards realisability, machine learning approaches remain at the forefront.

Few-shot fine-grained image classification (FSFGIC) methods refer to machine learning methods which aim to classify images (e.g., bird species, flowers, and airplanes) belonging to subordinate object categories of the same entry-level category with only a few samples. It is worth to note that feature representation learning is used not only to represent training samples, but also to construct classifiers for performing various FSFGIC tasks. In this paper, starting from the definition of FSFGIC, a taxonomy of feature representation learning for FSFGIC is proposed. According to this taxonomy, we discuss key issues on FSFGIC (including data augmentation, local or/and global deep feature representation learning, class representation learning, and task specific feature representation learning). The existing popular datasets and evaluation standards are introduced. Furthermore, a novel classification performance evaluation mechanism is designed with a 0.95 confidence interval for judging whether the classification accuracy obtained by a certain specified method is good or bad. Moreover, current challenges and future trends of feature representation learning on FSFGIC are elaborated.

Abstract: Botnets pose a grave cybersecurity threat, enabling widescale malicious activities through networks of compromised devices. Detecting botnets is challenging given their frequent use of evasion techniques like encryption. Traditional signature-based methods fail against modern botnets capable of zero-day attacks. This paper surveys recent advances applying machine learning for botnet detection based on analysis of network traffic payloads, flows, DNS data, and hybrid feature fusion. Core machine learning models include support vector machines, neural networks, random forests, and deep learning architectures, which extract patterns to separate benign and botnet behaviors automatically. Results demonstrate machine learning's capabilities in identifying heterogeneous botnets using artefacts in network streams. However, challenges remain around limited labeled data, real-time streaming, adversarial evasion, and model interpretability. Promising directions involve semi-supervised learning, adversarial training, scalable analytics, and explainable AI to address these gaps. Beyond the technical aspects, responsible development and deployment of botnet detection systems raise ethical considerations around privacy, transparency, and accountability. With diligent cross-disciplinary collaboration, machine learning promises enhanced, generalizable, and trustworthy techniques to combat the serious threat posed by continuously evolving botnets across the digital ecosystem.

The development of plant-based meat analogues has become a significant challenge for the food industry in recent years due to the increasing demand for sustainable and healthier proteins in the context of a global protein transition. Plant-based meat analogues imitate the visual, textural, and chemical properties of traditional meat products and are required to closely resemble meat to appeal to consumers. In addition, consumers demand natural, clean-label, and nutritional and healthy products. To address these challenges, the food industry must develop highly healthy, nutritious, and E-number-free meat analogue products. Understanding the functionality of each ingredient and its role in the food matrix is crucial to being a key player in the innovation of the meat analogues market. This review provides updated information on the primary ingredients utilized for the development of plant-based burger meat alternatives and their functionality. The key components of meat analogue burgers are outlined, including plant proteins, binding agents, fats and oils, flavorings, colorings, preservatives, fortificants, and clean-label considerations.

This paper presents the PEnsemble 4 model, a sophisticated machine learning framework that integrates IoT-based environmental data to accurately forecast maize yield. With the projected significant growth in global maize demand over the next decade, the inherent risks posed by the crop’s dependence on weather conditions necessitate improved prediction capabilities. The PEnsemble 4 model, developed with high accuracy, incorporates comprehensive datasets encompassing soil attributes, nutrient composition, weather conditions, and UAV-captured vegetation imagery. By employing a combination of Huber and M estimates, the PEnsemble 4 model effectively analyzes temporal patterns in vegetation indices, specifically CIre and NDRE, which serve as reliable indicators of canopy density and plant height. In addition, this research significantly contributes to precision agriculture by offering an efficient and sustainable alternative to conventional farming practices through precise yield predictions. Notably, the PEnsemble 4 model enables earlier estimation, advancing the timeline for yield prediction from the conventional day 100 in the R6 stage to day 79 in the R2 stage. This improvement enhances decision-making processes in farming operations. The remarkable accuracy rate of 91% underscores the importance of adopting a multifaceted data approach that harnesses IoT-derived environmental insights. Additionally, the PEnsemble 4 model extends its benefits beyond yield prediction, facilitating the detection of water and crop stress, as well as disease monitoring in broader agricultural contexts. Ultimately, the PEnsemble 4 model establishes a new standard in maize yield prediction, revolutionizing crop management and protection through the synergistic utilization of IoT and machine learning technologies.

Many complex behaviours in biological systems emerge from large populations of interacting molecules or cells, generating functions that go beyond the capabilities of the individual parts. Such collective phenomena are of great interest to bioengineers due to their robustness and scalability. However, engineering emergent collective functions is difficult because they arise as a consequence of complex multi-level feedback, which often spans multiple length-scales. Here, we present a perspective on how some of these challenges could be overcome by using multi-agent modelling as a design framework within synthetic biology. Using case studies covering the construction of synthetic ecologies to biological computation and synthetic cellularity, we show how multi-agent modelling can capture the core features of complex multi-scale systems and provide novel insights into the underlying mechanisms which guide emergent functionalities across scales. The ability to unravel design rules underpinning these behaviours offers a means to take synthetic biology beyond single molecules or cells and towards the creation of systems with functions that can only emerge from collectives at multiple scales.

In the realm of educational technology, Data Analytics (DA) is reshaping teaching and learning. This paper investigates the impact of DA on education, specifically comparing Student Information Systems (SIS) and Learning Management Systems (LMS). Data-driven insights from SIS and LMS platforms enable educators to tailor instructional strategies to individual student needs effectively. While SIS manages student data for administrative tasks, LMS offers tools for content delivery and collaboration. Using the strengths of SIS and LMS, educators can optimise teaching and learning experiences, tracking student progress, and fostering collaboration. The comparison emphasises their distinct roles: SIS manages administrative tasks, while LMS focusses on content delivery and collaboration. Using both systems, educators improve the teaching and learning experience, leveraging DA for continuous improvement. In summary, this paper underscores the transformative role of DA in education, framed within the context of comparing the SIS and LMS functionalities. By examining the intersection of data-driven decision making and educational technology, educators are equipped with the tools and knowledge to harness the power of data for meaningful and impactful teaching and learning.

In this paper, various machine learning techniques are discussed. These algorithms are used for many applications which include data classification, prediction, or pattern recognition. The primary goal of machine learning is to automate human assistance by training an algorithm on relevant data. This paper should also serve as a collection of various machine learning terminology for easy reference.

The current design of reinforcement learning methods demands exhaustive computing. Algorithms such as Deep Q-Network achieved outstanding results in the development of the area. However, the need for thousands of parameters and training episodes is still a problem. Thus, this document proposes a comparative analysis of the Q-Learning algorithm (the inception to create Deep Q Learning) and our proposed method termed M-Learning. The comparison among algorithms using Markov decision processes with delayed reward as a general testbench framework. Firstly, a full description of the main problems related to implementing Q-Learning, mainly about its multiple parameters. Then, the foundations of our proposed heuristic with its formulation and the whole algorithm were reported in detail. Finally, the methodology chosen to compare both algorithms was to train the algorithms in the Frozen Lake environment. The experimental results and an analysis of the best solutions found that our proposed algorithm highlights the differences in the number of episodes necessary and their standard variations. The code will be available on a GitHub repository once the paper is published.

In this paper, we present a novel approach for text independent phone-to-audio alignment based on phoneme recognition, representation learning and knowledge transfer. Our method leverages a self-supervised model (wav2vec2) fine-tuned for phoneme recognition using a Connectionist Temporal Classification (CTC) loss, a dimension reduction model and a frame-level phoneme classifier trained thanks to forced-alignment labels (using Montreal Forced Aligner) to produce multi-lingual phonetic representations, thus requiring minimal additional training. We evaluate our model using synthetic native data from the TIMIT dataset and the SCRIBE dataset for American and British English, respectively. Our proposed model outperforms the state-of-the-art (charsiu) in statistical metrics and has applications in language learning and speech processing systems. We leave experiments on other languages for future work but the design of the system makes it easily adaptable to other languages.

In the intersection of the Internet of Things (IoT) and Machine Learning (ML), the choice between high-level and low-level programming libraries presents a significant dilemma for developers, impacting not only the efficiency and effectiveness of ML models but also their environmental footprint. We have proposed a comprehensive framework to aid in this decision-making process, underpinned by a detailed comparative analysis of both types of libraries on one of the key IoT ML task; image classification. We have introduced a novel algorithm designed to calculate the green footprint of ML model training, factoring in execution time, memory utilization, power consumption, and CPU temperature, addressing the urgent need for sustainable ML practices. Through an empirical evaluation of popular libraries such as PyTorch for high-level and Libtorch for low-level development, we have assessed their performance, development efficiency, and hardware compatibility. The culmination of our research is a decision support system that synthesizes the experimental findings to guide developers toward choices that harmonize model performance with environmental sustainability.

Despite significant advances in tumor biology and clinical therapeutics, metastasis remains the primary cause of cancer-related deaths. While RNA-seq technology has been used extensively to study metastatic cancer characteristics, challenges persist in acquiring adequate transcriptomic data. To overcome this challenge, we propose MetGen, a generative contrastive learning based on deep learning model. MetGen generates synthetic metastatic cancer expression profiles using primary cancer and normal tissue expression data. Our results demonstrate that MetGen generates comparable samples to actual metastatic cancer samples, and we discuss the learning mechanism of MetGen. Additionally, we demonstrate MetGen's interpretability using metastatic prostate cancer and metastatic breast cancer. MetGen has learned highly relevant signatures in cancer, tissue, and tumor microenvironment, such as immune response and the metastasis process, which potentially fosters a more comprehensive understanding of metastatic cancer biology. The development of MetGen represents a significant step toward the study of metastatic cancer biology by providing a generative model that identifies candidate therapeutic targets for the treatment of metastatic cancer.

Introduction: Dental students’ use of online material to supplement learning has been studied, but educator awareness and reaction to the findings of this research is unknown. This small local study aimed to investigate dental students’ use of online content as a learning tool from an educator’s perspective.Methods: Educators at an Irish dental school completed a survey based on dental students’ use of online learning. Quantitative descriptive analyses were carried out as appropriate on the data collected, and a subsequent focus group was held to gain deeper insight into survey results. Data were analysed for themes by deductive and inductive coding methods.Results: This study found a sample of educators were aware of some student behaviours around online videos. Most educators correctly thought that students are likely to refer to online videos to prepare for dental procedures. Several educators believed that teachers should incorporate videos into their learning, but do not themselves upload or recommend online videos. Most recalled discussing accuracy and/or relevancy of online content with their students. Fewer thought that students would discuss contradictory videos with them. Focus group participants expressed concern over the accuracy of online content. This along with a lack of time were some reasons deterring them from referring students to online videos. Conclusions: An opportunity exists for dental educators and their students to have local dialogue around (1) student use, critical appraisal, and sharing of dental procedural video material; and (2) perceived benefits and challenges of incorporating both school-produced and online videos into teaching and learning. Although arguably non-generalizable beyond this context, this study may inform future research and serve to spark conversations within schools regarding video use for teaching clinical dental procedures.

Sleep disorder is a disease that can be categorized as both an emotional and physical problem. It imposes several difficulties and problems, such as distress during the day, sleep-wake disorders, anxiety, and several other problems. Hence, the main objective of this research is to utilize the strong capabilities of machine learning in the prediction of sleep disorders. In specific, this research aims to meet three main objectives. These objectives are to identify the best regression model, the best classification model, and the best learning strategy that highly suits sleep disorder datasets. Considering two related datasets and several evaluation metrics that are related to the tasks of regression and classification, the results revealed the superiority of the MultilayerPerceptron, SMOreg, and KStar regression models compared with the other twenty-three regression models. Also, IBK, RandomForest, and RandomizableFilteredClassifier showed superior performance compared with other classification models that belong to several learning strategies. Finally, the Function learning strategy showed the best predictive performance among the six considered strategies in both datasets and with respect to the most evaluation metrics.

Measuring Marginal Reflex Distance-1 (MRD-1) is a crucial clinical tool used to evaluate the position of the eyelid margin in relation to the cornea. Traditionally, this assessment has been conducted manually by plastic surgeons, ophthalmologists, or trained technicians. However, with the advancements in Artificial Intelligence (AI) technology, there is a growing interest in the development of automated systems capable of accurately measuring MRD-1.In this context, we introduce novel MRD-1 measurement methods based on deep learning algorithms that can simultaneously capture images and compute the results. This prospective observational study involved 154 eyes of 77 patients aged over 18 years who visited Chungnam National University Hospital between September 1, 2023, and July 29, 2023. We collected four different MRD1 da-tasets from patients using three distinct measurement methods, each tailored to the individual patient. The mean MRD1 values, measured through the manual method using a penlight, the deep learning method, ImageJ analysis from RGB eye images, and ImageJ analysis from IR eye images in 56 eyes of 28 patients, were 2.64 ± 1.04 mm, 2.85 ± 1.07 mm, 2.78 ± 1.08 mm, and 3.07 ± 0.95 mm, respectively. Notably, the strongest agreement was observed between MRD1_deep learning (DL) and MRD1_IR (0.822, p < 0.01). In the Bland-Altman plot, the smallest difference was observed between MRD1_DL and MRD1_IR ImageJ, with a mean difference of 0.0611 and △LOA (limits of agreement) of 2.5162, which is the smallest among the other groups. In conclusion, this novel MRD1 measurement method, based on an IR camera and deep learning, demon-strates statistical significance and can be readily applied in clinical settings.

This review paper investigates the influence of gestures on foreign language (FL) vocabulary learning through a series of experiments conducted in our laboratory. The manipulation of the gesture-word relationship was a consistent factor across the studies. Firstly, we examined the impact of gestures on noun and verb learning. The results revealed that participants exhibited better learning outcomes when FL words were accompanied by congruent gestures compared to a no gesture condition. This suggests that gestures have a positive effect on FL learning when there is a meaningful connection between the words and the accompanying gestures. However, in general, the recall of words in conditions where gestures were incongruent or lacked meaning was lower than in the no gesture condition. This indicates that under certain circumstances, gestures may have a detrimental impact on FL learning. We analyzed these findings in terms of their implications for facilitating or interfering with FL acquisition. Secondly, we addressed the question of whether individuals need to physically perform the gestures themselves to observe the effects of gestures on vocabulary learning. To explore this, participants were divided into two experimental groups. In one group, participants learned the words by actively performing the gestures ("do" learning group), while the other group simply observed the gestures performed by others ("see" learning group). The processing of congruent gestures facilitated the recall of FL words in both the "see" and "do" learning groups. However, the interference effect associated with processing incongruent gestures was more pronounced in the "see" learning group than in the "do" learning group. Thus, the performance of gestures appears to mitigate the negative impact that gestures may have on the acquisition of FL vocabulary. In conclusion, our findings suggest that iconic gestures can serve as an effective tool for learning vocabulary in a FL, particularly when the gestures align with the meaning of the words. Furthermore, the active performance of gestures helps counteract the negative effects associated with inconsistencies between gestures and word meanings. Consequently, if a choice must be made, a FL learning strategy in which learners acquire words while making gestures congruent with their meaning would be highly desirable.

Deep reinforcement learning is one of the research hotspots in artificial intelligence and has been successfully applied in many research areas, however, the low training efficiency and high demand for samples are problems that limit the application To address these problems, a hierarchical episodic control model extending episodic memory to the domain of hierarchical reinforcement learning is proposed in this paper. The model is theoretically justified and employs a hierarchical implicit memory planning approach for counterfactual trajectory value estimation. Starting from the final step and recursively moving back along the trajectory, a hidden plan is formed within the episodic memory. Experience is aggregated both along trajectories and across trajectories, and the model is updated using a multi-headed backpropagation similar to bootstrapped neural networks. This model extends the parameterized episodic memory framework to the realm of hierarchical reinforcement learning and is theoretically analyzed to demonstrate its convergence and effectiveness. Experiments conducted in Four Room, Mujoco, and UE4-based active tracking , highlight that the hierarchical episodic control model effectively enhances training efficiency. It demonstrates notable improvements in both low-dimensional and high-dimensional environments, even in cases of sparse rewards.

Federated learning (FL) has become a burgeoning and attractive research area, which provides a collaborative training scheme for distributed data sources with privacy concerns. Most existing FL studies focus on taking unimodal data, such as images and text, as the model input and resolving the heterogeneity challenge, i.e., the non-identically distributed (non-IID) challenge caused by data distribution imbalance related to data labels and data amount. In real-world applications, data are usually described by multiple modalities. However, to the best of our knowledge, only a handful of studies have been proposed to improve the system performance by utilizing multimodal data. In this survey paper, we identify the significance of this emerging research topic – multimodal federated learning (MFL) and perform a literature review on the state-of-art MFL methods. Furthermore, we categorize multi-modal federated learning into congruent and incongruent multimodal federated learning based on whether all clients possess the same modal combinations. We investigate the feasible application tasks and related benchmarks for MFL. Lastly, we summarize the promising directions and fundamental challenges in this field for future research.

Sentiment analysis of public opinion expressed in social networks has been developed into various applications, especially in English. Hybrid approaches are potential models for reducing sentiment errors on increasingly complex training data. This paper aims to test some hybrid deep learning models' reliability in some domains' Vietnamese language. Our research questions are to determine whether it is possible to produce hybrid models that outperform the Vietnamese language. Hybrid deep sentiment-analysis learning models are built and tested on reviews and feedback of the Vietnamese language. The hybrid models outperformed the accuracy of Vietnamese sentiment analysis on Vietnamese datasets. It contributes to the growing body of research on Vietnamese NLP, providing insights and directions for future studies in this area.

The demand for explainable and transparent models increases with the continued success of reinforcement learning. In this article, we explore the potential of generating shallow decision trees (DT) as simple and transparent surrogate models for opaque deep reinforcement learning (DRL) agents. We investigate three algorithms for generating training data for axis-parallel and oblique DTs with the help of DRL agents ("oracles") and evaluate these methods on classic control problems from OpenAI Gym. The results show that one of our newly developed algorithms, the iterative training, outperforms traditional sampling algorithms, resulting in well-performing DTs that often even surpass the oracle from which they were trained. Even higher dimensional problems can be solved with surprisingly shallow DTs. We discuss the advantages and disadvantages of different sampling methods and insights into the decision-making process made possible by the transparent nature of DTs. Our work contributes to the development of not only powerful but also explainable RL agents and highlights the potential of DTs as a simple and effective alternative to complex DRL models.

In this paper, we have investigated the impact of an early learning assessment on students' motivation for improving their performance throughout the semester. An observation analysis was conducted on an entry level mechanical engineering course in which students are enrolled in during their first semester of engineering work. This study analyzes the effect that a first exam, with an average below a passing grade, has on student's outcome in the course. It was hypothesized that students were motivated to achieve their desired grade outcomes following inadequate performance on the first exam. This was investigated by diving into the results of the course and referencing initial performance to the remaining exam and assessment outcomes. Students were placed into grade bands ranging from 0 to 100 in 20% increments. Their results were tracked and it was shown that for the second mechanics exam, averages jumped 43.333%, 35.35%, and 30.055% for grade bands of 0 to 20, 20 to 40, and 40 to 60 respectively. Assessment grades increased as well with the remaining assessments being averaged to a score of 91.095%. Variables contributing to student performance came from both with-in and outside the classroom. Learning communities, material differentiation, and student and professor adaptation all contributed to the rise in performance. It was concluded that the internal and external variables acted in combination with one another to increase student dedication to achieve success.

Machine learning methods have widely been applied to detect anomalies in machine and cutting tool behavior during lathe or milling. However, detecting anomalies in the workpiece itself did not get the same attention by researchers. That is why in this article, the authors present a pub-licly available multivariate time series dataset which was recorded during milling of 16MnCr5. Due to artificially introduced, though realistic anomalies in the workpiece the dataset can be ap-plied for anomaly detection. By using a convolutional autoencoder as a first model good results in detecting the location of the anomalies in the workpiece were achieved. Furthermore, milling tools with two different diameters where used which led to a dataset eligible for transfer learn-ing. The objective of this article is to provide researchers with a real-world time series dataset of the milling process which is suitable for modern machine learning research topics like anomaly detection and transfer learning.

While machine learning models are powering more and more everyday devices, there is a growing need for explaining them. This especially applies to the use of Deep Reinforcement Learning in solutions that require security, such as vehicle motion planning. In this paper, we propose a method of understanding what the RL agent’s decision is based on. The method relies on conducting statistical analysis on a massive set of state-decisions samples. It indicates which input features have an impact on the agent’s decision and the relationships between decisions, the significance of the input features, and their values. The method allows us for determining whether the process of making a decision by the agent is coherent with human intuition and what contradicts it. We applied the proposed method to the RL motion planning agent which is supposed to drive a vehicle safely and efficiently on a highway. We find out that making such analysis allows for a better understanding agent’s decisions, inspecting its behavior, debugging the ANN model, and verifying the correctness of input values, which increases its credibility.

Globally, zoonotic diseases have been on the rise in recent years. Predictive modelling approaches have been successfully used in the literature to identify the underlying causes of these zoonotic diseases. We examine the latest research in the field of predictive modeling that verifies the growth of zoonotic pathogens and assesses the factors associated with their spread. The results of our survey indicate that popular mathematical models can successfully be used in modeling the growth rate of these pathogens under varying storage temperatures. Additionally, some of them are used for the assessment of the inactivation of these pathogens based on various conditions. Based on the results of our study, machine learning models and deep learning are commonly used to detect pathogens within food items and to predict the factors associated with the presence of the pathogens.

COVID-19 has a severe risk of spreading rapidly, the quick identification of which is essential. In this regard, chest radiology images have proven to be a practical screening approach for COVID-19 affected patients. This study proposes a deep learning-based approach using Densenet-121 to detect COVID-19 patients effectively. We have trained and tested our model on the COVIDx dataset and performed both 2-class and 3-class classification, achieving 96.49% and 93.71% accuracy, respectively. By successfully utilizing transfer learning, we achieve comparable performance to the state-of-the-art method while using 15x fewer model parameters. Moreover, we performed an interpretability analysis using Grad-CAM to highlight the most significant image regions at test time. Finally, we developed a website that takes chest radiology images as input and detects the presence of COVID-19 or pneumonia and a heatmap highlighting the infected regions. Source code for reproducing results and model weights are available.

Background: Risk of metastatic recurrence of breast cancer after initial diagnosis and treatment depends on the presence of a number of risk factors. Although most univariate risk factors have been identified using classical methods, machine-learning methods are also being conducted to tease out non-obvious contributors to a patient’s individual risk of developing late distant metastasis. Bayesian-network algorithms may predict not only risk factors but also interactions among these risks, which consequently lead to metastatic breast cancer. We proposed to apply a previously developed machine-learning method to predict risk factors of 5-, 10- and 15-year metastasis. Methods: We applied a previously validated algorithm named the Markov Blanket and Interactive risk factor Learner (MBIL) on the electronic health record (EHR)-based Lynn Sage database (LSDB) from the Lynn Sage Comprehensive Breast Cancer at Northwestern Memorial Hospital. This algorithm provided an output of both single and interactive risk factors of 5-, 10-, and 15-year metastasis from LSDB. We individually examined and interpreted the clinical relevance of these interactions based on years to metastasis and the reliance on interactivity between risk factors. Results: We found that with lower alpha values (low interactivity score), the prevalence of variables with an independent influence on long term metastasis was higher (i.e., HER2, TNEG). As the value of alpha increased to 480, stronger interactions were needed to define clusters of factors that increased the risk of metastasis (i.e., ER, smoking, race, alcohol usage). Conclusion: MBIL identified single and interacting risk factors of metastatic breast cancer, many of which were supported by clinical evidence. These results strongly recommend the development of further large data studies with different databases to validate the degree to which some of these variables impact metastatic breast cancer in the long term.

Many works have recently identified the need to combine deep learning with extreme learning to strike a performance balance with accuracy especially in the domain of multimedia applications. Considering this new paradigm, namely convolutional extreme learning machine (CELM), we present a systematic review that investigates alternative deep learning architectures that use extreme learning machine (ELM) for a faster training to solve problems based on image analysis. We detail each of the architectures found in the literature, application scenarios, benchmark datasets, main results, advantages, and present the open challenges for CELM. We follow a well structured methodology and establish relevant research questions that guide our findings. We hope that the observation and classification of such works can leverage the CELM research area providing a good starting point to cope with some of the current problems in the image-based computer vision analysis.

Hateful and abusive speech presents a major challenge for all online social media platforms. Recent advances in Natural Language Processing and Natural Language Understanding allow more accurate detection of hate speech in textual streams. This study presents a multimodal approach to hate speech detection by combining Computer Vision and Natural Language processing models for abusive context detection. Our study focuses on Twitter messages and, more specifically, on hateful, xenophobic and racist speech in Greek aimed at refugees and migrants. In our approach we combine transfer learning and fine-tuning of Bidirectional Encoder Representations from Transformers (BERT) and Residual Neural Networks (Resnet). Our contribution includes the development of a new dataset for hate speech classification, consisting of tweet ids, along with the code to obtain their visual appearance, as they would have been rendered in a web browser. We have also released a pre-trained Language Model trained on Greek tweets, which has been used in our experiments. We report a consistently high level of accuracy (accuracy score=0.970, f1-score=0.947 in our best model) in racist and xenophobic speech detection.

The exponential growth of biomedical data in recent years urged the application of numerous machine learning techniques to address emerging problems in biology and clinical research. By enabling automatic feature extraction, selection and generation of predictive models, these methods can be used to efficiently study complex biological systems. Machine learning techniques are frequently integrated with bioinformatic methods, as well as curated databases and biological networks, to enhance training and validation, identify the best interpretable features, and enable feature and model investigation. Here, we review recently developed methods that incorporate machine learning within the same framework with techniques from molecular evolution, protein structure analysis, systems biology and disease genomics. We outline the challenges posed for machine learning, and in particular, deep learning in biomedicine and suggest unique opportunities for machine learning techniques integrated with established bioinformatics approaches to overcome some of these challenges.

Education setting evolved from historical open learning system to traditional classroom set-up to distance learning modality. Teaching-Learning practice is transformed with an evolution of teaching-learning materials. With technological advancement in progressive manner and it’s increasing use in academic setting, distance learning has been the on-demand and on-debate topic in current educational discourse. Comparatively fresh topic in Nepali academic setting, this paper intended to analyze the perception of Nepali students towards online modality in Nepali academic setting. This paper further analyzed the student’s preference towards distance learning in current Nepali academic setting. Research findings were analyzed based on data collected through literature review, interview with students and professor and quantitative data collection through use of google form. Study identified opportunities as revenue generation; continuation of academic career from any part of country; increase learning outcome among jobholders. Study identified challenges as unequal access and quality of internet facilities; affordability of laptops/computers; limited interaction; and frequent disturbances. Seeing the better prospects, study strongly supported the need of shift in academic shift from traditional setting to non-traditional setting in Nepali context to meet the global needs of competitive and quality education.

Identifying dust aerosols from passive satellite images is of great interest for many applications. In this study, we developed 5 different machine-learning (ML) and deep-learning (DL) based algorithms, including Logistic Regression, K Nearest Neighbor, Random Forest (RF), Feed Forward Neural Network (FFNN), and Convolutional Neural Network (CNN), to identify dust aerosols in the daytime satellite images from the Visible Infrared Imaging Radiometer Suite (VIIRS) under cloud-free conditions on a global scale. In order to train the ML and DL algorithms, we collocated the state-of-the-art dust detection product from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) with the VIIRS observations along the CALIOP track. The 16 VIIRS M-band observations with the center wavelength ranging from deep blue to thermal infrared, together with solar-viewing geometries and pixel time and locations, are used as the predictor variables. Four different sets of training input data are constructed based on different combinations of VIIRS pixel and predictor variables. The validation and comparison results based on the collocated CALIOP data indicates that the FFNN method based on all available predictor variables is the best performing one among all methods. It has an averaged dust detection accuracy of about 81 %, 89 % and 85 % over land, ocean and whole globe, respectively, compared with collocated CALIOP. When applied to off-track VIIRS pixels, the FFNN method retrieves geographical distributions of dust that are in good agreement with on-track results as well as CALIOP statistics. For further evaluation, we compared our results based on the ML and DL algorithms to NOAA’s Aerosol Detection Product (ADP) , which is a product that classifies dust, smoke and ash using physical-based methods. The comparison reveals both similarity and differences. Overall, this study demonstrates the great potential of ML and DL methods for dust detection and proves that these methods can be trained on the CALIOP track and then applied to the whole granule of VIIRS granule.

Deep learning machines are computational models composed of multiple processing layers of adaptive weights to learn representations of data with multiple levels of abstraction. Their structure is mainly reflecting the intuitive plausibility of decomposing a problem into multiple levels of computation and representation since it is believed that higher layers of representation allow a system to learn complex functions. Surprisingly, after decades of research, from learning and design perspectives these models are still deployed in a heuristic manner. In this paper, deep learning feed-forward machines are modeled from a statistical mechanics point of view as disordered physical systems where its macroscopic behavior is determined in terms of the interactions defined between the basic constituent of these models, namely, the artificial neuron. They are viewed as the equilibrium states of a theoretical body that is subject to the law of increase of the entropy. The study of the changes in energy of the body when passing from one equilibrium state to another is used to understand the structure and role of the phase space of the system, the stability of the equilibrium states, and the resulting degree of disorder. It is shown that the topology of these models is strongly linked to their stability and resulting level of disorder. Furthermore, the proposed theoretical characterization permit to assess the thermodynamic efficiency with which information can be processed by these models, and to provide a practical methodology to quantitatively estimate and compare their expected learning and generalization capabilities. These theoretical results provides new insights to the theory of deep learning and their implications are shown to be consistent through a set of benchmarks designed to experimentally assess their validity.

In the Agriculture sector, control of plant leaf diseases is crucial as it influences the quality and production of plant species with an impact on the economy of any country. Therefore, automated identification and classification of plant leaf disease at an early stage is essential to reduce economic loss and to conserve the specific species. Previously, to detect and classify plant leaf disease, various Machine Learning models have been proposed; however, they lack usability due to hardware incompatibility, limited scalability and inefficiency in practical usage. Our proposed DeepLens Classification and Detection Model (DCDM) approach deal with such limitations by introducing automated detection and classification of the leaf diseases in fruits (apple, grapes, peach and strawberry) and vegetables (potato and tomato) via scalable transfer learning on A.W.S. SageMaker and importing it on AWS DeepLens for real-time practical usability. Cloud integration provides scalability and ubiquitous access to our approach. Our experiments on extensive image data set of healthy and unhealthy leaves of fruits and vegetables showed an accuracy of 98.78% with a real-time diagnosis of plant leaves diseases. We used forty thousand images for the training of deep learning model and then evaluated it on ten thousand images. The process of testing an image for disease diagnosis and classification using AWS DeepLens on average took 0.349s, providing disease information to the user in less than a second.

Using machine learning or deep learning to solve the problem of garbage recognition and classification is an important application in computer vision, but due to the incomplete establishment of garbage datasets and the poor performance of complex network models on smart terminal devices, the existing garbage classification models The effect is not good.This paper presents a waste classification and identification method base on transfer learning and lightweight neural network. By migrating the lightweight neural network MobileNetV2 and rebuild it, The reconstructed network is used for feature extraction, and the extracted features are introduced into the SVM to realize the identification of 6 types of garbage. The model was trained and verified by using 2527 pieces of garbage labeled data in the TrashNet dataset, which ultimately resulted in a classification accuracy of 98.4% of the method, which proves that the method can effectively improve the classification accuracy and time and overcome the problem of weak data and less labeling. The over-fitting phenomenon encountered by small data sets in deep learning makes the model robust.

In this brief letter we report our initial results on the application of deep-learning to the massive MIMO channel estimation challenge. We show that it is possible to estimate wireless channels and that the possibility of mitigating pilot-contamination with deep-learning is possible given that the leaning model underwent an extensive training-phase and that it has been presented with a large number of different channel conditions.

If students do not fully apply themselves, then they may be considered responsible for the result of being inadequately prepared. +- Nevertheless, student outcomes are more likely to reflect a combination of both effort and systematic problems with overall course architecture. Deficiencies in course design result in inadequate preparation that adversely and directly impacts students’ productivity upon entering the workforce. Such an impact negatively influences students' ability to maintain gainful employment and provide for their families, which inevitably contributes to the development of issues concerning their psychological well-being. It is well-documented that incorporating active learning strategies in course design and delivery can enhance student learning outcomes. Despite the benefit of implementing active learning techniques, rarely in the real world will it be possible for techniques to be used in isolation of one another. Therefore, the purpose of this proposed study is to determine the interactive effects of two active learning strategies because, at a minimum, technique-pairs more accurately represent the application of active learning in the natural educational setting. There is a paucity of evidence in the literature directed toward investigating the interactive effects of multiple active learning techniques that this study is aimed at filling. The significance of this research is that, by determining the interactive effects of paired active learning strategies, other research studies on the beneficial effects of using particular active learning technique-pairs will be documented contributing to the literature so that ultimately classroom instruction may be customized according to the determination of optimal sequencing of strategy-pairs for particular courses, subjects, and desired outcomes that maximize student learning.

In the wake of increasing digital fraud, this paper introduces an innovative application of Artificial Intelligence (AI) in detecting fraudulent activities across finance, healthcare, and e-commerce sectors. It presents a detailed analysis of machine learning methodologies, specifically focusing on the advantages of supervised, unsupervised, and deep learning techniques. The paper addresses the challenges such as data imbalance, model interpretability, and ethical implications in AI-based fraud detection. It also discusses the necessity of high-quality datasets and advocates for the integration of traditional and advanced machine learning methods to enhance accuracy and adaptability in fraud identification. However, it acknowledges the limitations including computational demands and overfitting risks. The study underscores the importance of collaborative efforts between AI experts and industry professionals to develop ethical, efficient, and reliable AI solutions for fraud detection.

Text annotation is an essential element of the natural language processing approaches. The manual annotation process performed by humans has several drawbacks, such as subjectivity, slowness, fatigue, and possibly carelessness. In addition, annotators may annotate ambiguous data. So, we developed the concept of automated annotation to get the best annotations using several machine-learning approaches. The proposed approach is based on an ensemble algorithm of meta-learners and meta-vectorizer techniques. The approach employs a semi-supervised learning technique for automated annotation, aimed at detecting hate speech. This involves leveraging various machine learning algorithms, including Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbors (KNN), and Naive Bayes (NB), in conjunction with Word2Vec and TF-IDF text extraction methods. The annotation process is performed using 13,169 Indonesian YouTube comments data. The proposed model used a Stemming approach using data from Sastrawi and also new data of 2,245 words. Semi-supervised learning uses 5%, 10%, and 20% of labeled data as compared to performing labeling based on 80% of the datasets. In semi-supervised learning, the model learns from the labeled data, which provides explicit information, and the unlabeled data, which offers implicit insights. This hybrid approach enables the model to generalize and make informed predictions even when limited labeled data is available, ultimately enhancing its ability to handle real-world scenarios with scarce annotated information. In addition, the proposed method uses a variety of thresholds for matching words labeled with hate speech ranging from 0.6, 0.7, 0.8, and 0.9. The experiment showed that the KNN-Word2ec model has the best accuracy value of 96.9% with a scenario of 5%:80%:0.9. However, several other methods have also accuracy above 90%, such as SVM and DT based on both text extraction methods in several test scenarios.

Plasma technology shows tremendous potential for revolutionizing oncology research and treatment. Reactive oxygen and nitrogen species, electromagnetic emissions generated through gas plasma jets, have attracted significant attention due to their selective cytotoxicity towards cancer cells. To leverage the full potential of plasma medicine, researchers have explored the use of mathematical models and various subsets of machine learning, such as reinforcement learning, and deep learning. This review emphasizes the significant application of AI algorithms in the adaptive plasma system, paving the way for precision and dynamic cancer treatment. Realizing the full potential of AI in plasma medicine, requires research efforts, data sharing and interdisciplinary collaborations. Unravelling the complex mechanisms, developing real-time diagnostics, and optimizing AI models will be crucial to harness the true power of plasma technology in oncology. The integration of personalized and dynamic plasma therapies, alongside AI and diagnostic sensors, presents a transformative approach to cancer treatment with the potential to improve outcomes globally.

Higher education teaching staff members need to build a scientifically accurate and comprehensive understanding of the function of the brain in learning to optimize teaching and achieve excellent student learning. An international consortium developed a professional development six-module course on educational neuroscience and online community of practice applying design thinking. A mixed methods research design was employed to investigate the attitudes of thirty-two (N=32) participating academics using a survey comprising eleven closed and open questions. Data analysis methods included descriptive statistics, correlation, generalized additive model and grounded theory. The overall evaluation demonstrated a notable satisfaction level with regard to the quality of the course. Given the power of habits, mentoring and peer interactions are recommended to ensure the effective integration of theoretical neuroscientific evidence into teaching practice.

Attitude control of a novel regional truss-braced wing aircraft with low stability characteristics is addressed in this paper using Reinforcement Learning (RL). In recent years, RL has been increasingly employed in challenging applications, particularly, autonomous flight control. However, a significant predicament confronting discrete RL algorithms is the dimension limitation of the state-action table and difficulties in defining the elements of the RL environment. To address these issues, in this paper, a detailed mathematical model of the mentioned aircraft is first developed to shape an RL environment. Subsequently, Q-learning, the most prevalent discrete RL algorithm will be implemented in both the Markov Decision Process (MDP), and Partially Observable Markov Decision Process (POMDP) frameworks to control the longitudinal mode of the air vehicle. In order to eliminate residual fluctuations that are a consequence of discrete action selection, and simultaneously track variable pitch angles, a Fuzzy Action Assignment (FAA) method is proposed to generate continuous control commands using the trained Q-table. Accordingly, it will be proved that by defining an accurate reward function, along with observing all crucial states (which is equivalent to satisfying the Markov Property), the performance of the introduced control system surpasses a well-tuned Proportional–Integral–Derivative (PID) controller.

One of the most important challenges in the Machine and Deep Learning areas today is to build good models using small datasets, because sometimes it is not possible to have large ones. Several techniques have been proposed in the literature to address this challenge. This paper aims at studying the different available Deep Learning techniques and performing a thorough experimentation to analyze which technique or combination thereof improves the performance and effectiveness of the models. A complete comparison with classical Machine Learning techniques was carried out, to contrast the results obtained using both techniques when working with small datasets. Thirteen algorithms were implemented and trained using three different small datasets (MNIST, Fashion MNIST, and CIFAR-10). Each experiment was evaluated using a well-established set of metrics (Accuracy, Precision, Recall, F1, and the Matthews correlation coefficient). The experimentation allowed concluding that it is possible to find a technique or combination of them to mitigate a lack of data, but this depends on the nature of the dataset, the amount of data, and the metrics used to evaluate them.

Intertidal seagrass plays a vital role in estimating the overall health and dynamics of coastal environments due to its interaction with tidal changes. However, most seagrass habitats around the globe have been in steady decline due to human impacts, disturbing the already delicate balance in environmental conditions that sustain seagrass. Miniaturization of multi-spectral sensors has facilitated very high resolution mapping of seagrass meadows, which significantly improve the potential for ecologists to monitor changes. In this study, two analytical approaches used for classifying intertidal seagrass habitats are compared: Object-based Image Analysis (OBIA) and Fully Convolutional Neural Networks (FCNNs). Both methods produce pixel-wise classifications in order to create segmented maps, however FCNNs are an emerging set of algorithms within Deep Learning with sparse application towards seagrass mapping. Conversely, OBIA has been a prominent solution within this field, with many studies leveraging in-situ data and multiscale segmentation to create habitat maps. This work demonstrates the utility of FCNNs in a semi-supervised setting to map seagrass and other coastal features from an optical drone survey conducted at Budle Bay, Northumberland, England. Semi-supervision is also an emerging field within Deep Learning that has practical benefits of achieving state of the art results using only subsets of labelled data. This is especially beneficial for remote sensing applications where in-situ data is an expensive commodity. For our results, we show that FCNNs have comparable performance with standard OBIA method used by ecologists, while also noting an increase in performance for mapping ecological features that are sparsely labelled across the study site.

The COVID-19 outbreaks have caused universities all across the globe to close their campuses and forced them to initiate online teaching. This article reviews the pedagogical foundations for developing effective distance education practices, starting from the assumption that promoting autonomous thinking is an essential element to guarantee full citizenship in a democracy and for moral decision making in situations of rapid change, which has become a pressing need in the context of a pandemic. In addition, the main obstacles related to this new context are identified, and solutions are proposed according to the existing bibliography in learning sciences.

In this article, we analyse the computer aid screening method of COVID19 using Xray and CT scan images. The main objective is to set an analytical closure about the computer aid screening of COVID19 among the X-ray image and CT scan image. The computer aid screening method includes deep feature extraction, transfer learning and traditional machine learning image classification approach. The deep feature extraction and transfer learning method considered 13 pre-trained CNN model. The machine learning approach includes three sets of features and three classifiers. The pre-trained CNN models are alexnet, googlenet, vgg16, vgg19, densenet201, resnet18, resnet50, resnet101, inceptionv3, inceptionresnetv2, xception, mobilenetv2 and shufflenet. The features and classifiers in machine learning approaches are GLCM, LBP, HOG and KNN, SVM, Naive bay’s respectively. In addition, we also analyse the different paradigms of classifiers. In total, the comparative analysis is carried out in 65 classification models, i.e. 13 in deep feature extraction, 13 in transfer learning and 39 in machine learning approaches. Finally, all the classification models perform better in X-ray image set compare to CT scan image set.

(1) Background: Evidence-based policymaking requires data about the local population's socioeconomic status (SES) at detailed geographical level, however, such information is often not available, or is too expensive to acquire. Researchers have proposed solutions to estimate SES indicators by analyzing Google Street View images, however, these methods are also resource-intensive, since they require large volumes of manually labeled training data. (2) Methods: We propose a methodology for automatically computing surrogate variables of SES indicators using street images of parked cars and deep multiple instance learning. Our approach does not require any manually created labels, apart from data already available by statistical authorities, while the entire pipeline for image acquisition, parked car detection, car classification, and surrogate variable computation is fully automated. The proposed surrogate variables are then used in linear regression models to estimate the target SES indicators. (3) Results: We implement and evaluate a model based on the proposed surrogate variable at 30 municipalities of varying SES in Greece. Our model has $R^2=0.76$ and a correlation coefficient of $0.874$ with the true unemployment rate, while it achieves a mean absolute percentage error of $0.089$ and mean absolute error of $1.87$ on a held-out test set. Similar results are also obtained for other socioeconomic indicators, related to education level and occupational prestige. (4) Conclusions: The proposed methodology can be used to estimate SES indicators at the local level automatically, using images of parked cars detected via Google Street View, without the need for any manual labeling effort.

This study investigates the latest advancements in utilizing machine learning techniques for predicting financial market movements. Financial markets are complex systems influenced by numerous factors, making accurate predictions challenging but highly valuable. The research explores various machine learning approaches, including supervised learning, unsupervised learning, reinforcement learning, and deep learning, to enhance the accuracy and reliability of financial market forecasts. The study provides a comprehensive review of current methodologies, highlighting key innovations and their applications in financial market prediction. Techniques such as neural networks, support vector machines, random forests, and ensemble methods are examined in the context of their ability to analyze vast amounts of financial data, identify patterns, and generate actionable insights. Additionally, the research delves into the integration of alternative data sources, such as social media sentiment, news articles, and macroeconomic indicators, to improve prediction models. Through empirical analysis and case studies, the study demonstrates the effectiveness of advanced machine learning models in predicting stock prices, currency exchange rates, and commodity prices. It also discusses the challenges associated with these techniques, including overfitting, data quality issues, and the need for interpretability in model outputs. Findings suggest that machine learning techniques can significantly enhance the predictive power of financial models, offering more accurate and timely insights for investors and financial analysts. The study concludes with recommendations for future research directions, emphasizing the importance of model transparency, the incorporation of diverse data sets, and the continuous adaptation of machine learning algorithms to evolving market conditions.

Several years now, the global scientific community has accepted and recognized the importance of Evidence-based practice (EBP) for Nursing science. The main factor for the implementation of EBP is the competence of undergraduate Nursing students towards EBP, so that they as active nurses be ready for its application in their clinical practice, in order to provide better care for their patients. The aim of the present study is to examine the level of evidence-based practice competency evaluated with the self-reported EBP-COQ questionnaire. A quantitative study based on cross-sectional design was conducted from February to June 2022 in a convenience sample of Greek undergraduate nursing students. The Evidence-Based Practice Competence Questionnaire (EBP-COQ) and a questionnaire on undergraduate students’ gender, age, academic year, and training in the field of research were used. The SPSS 26.0 program was used to perform descriptive, bivariate and multivariate analyses. A total of 175 undergraduate students participated at the Hellenic University Nursing (Greece), specifically from the 2nd and 3rd academic year. The mean score of EBP-COQ for nursing undergraduate students was 3.03 ± 0.26, indicating moderate EBP. Among the EBP-COQ dimensions, the mean scores were obtained 3.03 ± 0.32 for Attitude toward EBP, 3.01 ± 0.49 for Skills in EBP, and 3.03 ± 0.49 for Knowledge in EBP. They were not found significant differences among participants’ mean EBP-COQ scores regarding gender (p = 0.766), age (p = 0.400), academic year (p = 0.153), and training in the field of Research Methodology (p = 0.538). It appears that the level of readiness towards EBP is mediocre between undergraduate nursing students at a university in Greece. Therefore, it is necessary to carry out new studies in the future, so that there is a correct approach of all the elements that contribute to the readiness of Nursing undergraduate students regarding EBP.

The implementation and sustainability of evidence-based practice (EBP) is a complex process. Best practice guidelines (BPGs) can facilitate the translation of knowledge from theory into practice, but they require working groups to adapt recommendations and encourage adherence to them. The aim of this study was to identify facilitators and barriers of BPGs in the Registered Nurses’ Association of Ontario®, within the framework of Best Practice Spotlight Organisations® (BPSOs®), in a hospital setting. The study was conducted using the modified Delphi method (two rounds) with a cohort of BPSO® Best Practice Champions. The facilitators and barriers included in the first round were identified through a bibliographic search. The degree of concordance was determined in the second round using the highest mean resulting values. A total of 44 nurses responded, with a mean age of 42.57±8.563. Facilitators included: Work Environment; Working in a hospital that encourages research; and Dedicating time to EBP (p&lt;0.001). Barriers: Excessive workload; Professional mobility (p&lt;0.001); and Lack of time (p=0.002). As such, it is necessary to approach human resource management by implementing new policies that guarantee systemic sustainability. The BPSO® programme is an excellent framework for promoting the translation of theoretical knowledge into practice and effecting institutional change.

This article explores the transformative potential of fermentation in elevating the quality of plant-based matrices to match the desirable attributes of traditional dairy and meat products. As the demand for sustainable products without animal welfare issues increases, fermentation emerges as a key process to enhance the organoleptic properties and nutritional content of plant-based analogs. This study explores the effect of fermentation when applied to legume matrices, focusing on the resulting texture, flavor, and nutritional value. A selection of Bacillus subtilis, lactic acid bacteria (LAB) strains, and combinations thereof, showed potential for improving the aforementioned organoleptic and nutritional characteristics of fermented plant bases. In four different legume-derived matrices, fermentation improved texture, degraded undesirable plant carbohydrates, and removed off-flavor compounds, while producing desirable dairy-associated compounds. The degradation of the undesirable beany off-flavor-causing compound hexanal appears to be a universal phenomenon, as every tested strain as well as their combinations exhibited the capability to decrease the hexanal content albeit with varying efficiency. Some LAB strains were found to be capable of producing carotenoids and might hence have the potential for tailoring fermented plant-based matrices for specific applications, such as yellow cheese or red meat analogs.

Direct competitions involve competitors that mutually contend for the same resource or objective. A controlled, well-documented, and data-rich field involving direct competition is represented by sports. Usually the winner of sports competitions is believed to be the one with the greatest talent, however, there are other factors that contribute to the outcome of a competition, there are in fact random, unpredictable events that can change the outcome of a competition. For this reason, if one wants to understand the properties of a competition is necessary to have a method to evaluate the importance of the talent or aptitude of a competitor, as opposed to the importance of chance, in determining the outcome of the competition itself. In this work we study the sport of tennis using the data obtained from the Association of Tennis Professionals (ATP) tournaments. We construct an agent-based model that is able to produce data analogous to the real one; this model depends on three parameters, the talent, the action of chance, and the weight of talent. In particular, we don't fix the values of these parameters and we fit the model results using a genetic algorithm, in this way, we study all the possible combinations of parameters, in the parameter space, that are able to reproduce the real systems data. We show that the model fits well the real data only for limited regions of the parameter space. On these limited regions of the parameter space are possible further optimization of the model results, limiting the values of parameters. In this way, our agent-based model, by means of this genetic algorithm calibration, is able to provide us with useful information without any a priori constraints.

Patients with Multiple Myeloma have an increased susceptibility to severe infections due to several various factors, especially the high tumor burden. The use of the anti-CD38 monoclonal antibodies-based therapies has changed the frequency and epidemiology of infections. We report herein the safety profile of the pivotal clinical trials conducted with anti-CD38 mAbs (daratumumab and isatuximab) and anti-SLAMF7 mAbs (elotuzumab) that led to the approval of regimens employed today in NDMM and RRMM patients.

In this study we evaluate performance of two American surveillance systems for monitoring the spread of antimicrobial-resistant (AMR) gonorrhea: the Gonococcal Isolate Surveillance Project (GISP) and the enhanced Gonococcal Isolate Surveillance Project (eGISP) which includes the non-urethral isolates in addition to the urethral ones utilized in the original surveillance system. A continuous-time agent-based model of gonorrhea transmission among the US men who have sex with men (MSM) population was developed and used for this purpose. Our results show that the accuracy of eGISP was nearly 10% higher than the accuracy of GISP (97.8% (95% uncertainty interval: 83.5%, 100%) vs. 87.9% (55%, 100%)). Most of time, GISP detects the moment when the resistance to the first-line antibiotic reaches the switch threshold recommended by the WHO later than when it was actually reached. This leads to the additional spread of AMR gonorrhea. We also found that including the extragenital isolates in the surveillance is more important for the accurate monitoring than increasing the number of isolates. Finally, our results demonstrate that rectum and pharynx can serve as a niche for fostering antimicrobial resistance. These findings provide a scientific basis which can be used to inform the decision made by the policymakers on how many GISP and eGISP sites to establish each year.

This article evaluates the reliability and consistency of nutrition- and supplementation-related advice for pregnant women provided by ten selected health-related Internet portals. The portals were chosen based on their perceived reliability and prominence in Google searches, with representation from both English and Polish language sources. The evaluation criteria included the adherence of the presented information to official recommendations and its evidence-based character based on specific items representing dietary aspects important in pregnancy. While the overall reliability was deemed acceptable, significant variations existed both among the portals and specific evaluated items. Notably, HealthLine, Medline Plus, and NCEZ emerged as the most evidence-based, while WebMD and Medycyna Praktyczna were identified as less reliable. Despite a number of issues, the analysed portals remain valuable sources of nutritional information for pregnant women, offering user-friendly accessibility superior to alternatives such as social media, on the one hand, and scientific articles, on the other. Improved consistency and attention to detail, especially in relation to vitamin intake and supplementation, would improve the overall quality of health portals.

Approximately 20% of the world's population resides in coastal areas at risk of rising sea levels due to climate change. This could lead to severe weather events and ecological impact on critical infrastructure. To combat the effects of water-based environmental forces on the land, hard engineering methods such as seawalls have been utilized. However, a more sustainable solution is emerging in nature-based engineering, which involves growing vegetation. This approach can help extend the lifespan of existing technology and provide climate change adaptation and resilience for coastal and riverine communities. The following paper will review seaweed farms as an advanced nature-based mitigation approach. The paper also outlines the findings of experiments at RWTH Aachen University's hydraulic laboratory. These tests involved using seaweed as a natural solution to dampen waves in a model system to test a hypothesis. One result involving a system with two lines of seaweed revealed 15 percent wave damping. Using seaweed as a nature-based solution, a gentle engineering approach to designing future vegetated protection systems can extend the design life of existing coastal infrastructure. This can also provide protection against climate-induced sea level rise, promote coastal risk mitigation, restore ecosystems, and foster blue bio-economic development.

The digital era has significantly amplified the volume of online reviews, presenting both opportunities and challenges in harnessing insights from consumer feedback. Aspect-Based Sentiment Analysis (ABSA) has emerged as a crucial tool for distilling sentiments from these reviews, providing valuable data for enhancing product and service quality. This study introduces an advanced hybrid model, AdvSentiNet, which integrates adversarial training into the state-of-the-art framework to elevate the precision of sentiment detection at the aspect level. By employing an adversarial network, where a generative model competes against a classifier by crafting highly realistic synthetic samples, we aim to bolster the model's resilience against varied data samples. This innovative approach, unexplored in its entirety within the realm of ABSA, demonstrated remarkable performance improvements on benchmark datasets. For instance, accuracy on the SemEval 2015 dataset escalated from 81.7% to 82.5%, and for the SemEval 2016 dataset, it surged from 84.4% to 87.3%.

The interest in virtual fencing, as opposed to traditional physical barriers, has increased since it makes it easier to move the boundaries of animal enclosures for agricultural and rewilding purposes. Despite this, the implementation of virtual fence systems, such as the Nofence© system, requires large financial investments. The necessity of individual collars also makes it more difficult to implement this system on a larger scale. If a herd of cattle tend to maintain cohesive social groups around dominant individuals, it might be possible to use fewer collars thereby making the system more cost efficient. This study investigates the pattern in spatial distribution in a herd of 17 Angus cows on Fanø in Denmark with GPS locations, using a Nofence© system. The aim of this paper is to determine how individuals position themselves in a herd, spatially, and identify a pattern in ranks. The method used in this study examines the distances between an individual to the rest of the herd mates using nearest neighbour and distribution plots. Contrary to the expectations, this study did not show a distinct pattern in herd distribution. While some tendencies in spatial distribution patterns were observed, only a low concordance could be found (W=0.15,p=3.02e−16). Therefore, interactions among individuals should be considered in future studies, rather than only investigating the herd as one unit. A cumulative curves of the ranks estimated over the entire periods allowed however a rough estimation of the hierarchy and allowed the identification of the highest ranked cows. Therefore it might be possible to use fewer collars if the high ranked cows are discovered, but the estimation of the hierarchy will require a period of observation overall several days as the hierarchical positions are fluctuating too much from days to days. Therefore the cumulative rank curves seems to be a valid solution.

Hybrid materials are a recent addition in the field of Restorative Dentistry for computer-assisted design/ computer-assisted manufacturing (CAD/CAM) indirect restorations. The long–term clinical success of modern dental restorative materials follows a multifactorial pattern. Among the characteristics, affecting the longevity of a restoration, mechanical properties and physicο – chemical interactions are of utmost importance. While numerous researchers constantly evaluate the mechanical properties, the biological background of resin–based CAD/CAM biomaterials is scarcely investigated and, therefore, less described in the literature. This review aims to analyze the biofilm formation on the surfaces of novel hybrid, resin–based CAD/CAM materials and evaluate the methodological protocols followed to assess microbial growth. It is demonstrated that the surface structure, the composition and the finishing and polishing procedures on the surface of a dental restorative material influence the initial bacterial adhesion; however most studies focus on in vitro protocols, whereas in vivo and/or in situ research of microbiomics in CAD/CAM restorative materials is lacking, obstructing in that manner the accurate understanding of the bioadhesion phenomenon in the oral cavity.

Apoptotic cells possess immunomodulatory effects that can be utilized to treat imbalanced immune conditions. Information on the preclinical safety of such treatment is sparse. In this study, the safety of apoptotic cells (Allocetra-OTS) was assessed in a GLP toxicological study on Sprague Dawley rats. Three doses of Allocetra-OTS or vehicle were administered intravenously (IV) for 3 consecutive days. Animals in the Main study were sacrificed on day 4, while animals from the Recovery groups were kept for 14 or 28 days. Allocetra-OTS was well tolerated, and no adverse effects were observed in body weight, clinical signs, food consumption or ophthalmologic observation. Thus, the No Observed Adverse Effect Level (NOAEL) dose was determined as the highest dose administered. An observed elevation in immune cells was suspected to be due to Allocetra-OTS; similarly to other clinical chemistry parameters, however, it was resolved in the Recovery phases. Splenomegaly and a dose-related Extramedullary Hematopoiesis (EMH) in the red pulp were observed, with no adverse events, and were considered to be a normal and expected reaction following IV administration of cell-based therapies. In conclusion, under the conditions of this study Allocetra-OTS was concluded to be safe, further supporting its potential candidacy for clinical studies

Dysmenorrhea symptoms are frequent and often self-treated using non-prescription medicines. To further characterize women with dysmenorrhea using a combination of hyoscine butylbromide and paracetamol (PLUS) for self-management of their complaints, we performed a secondary analysis of a published pharmacy-based patient survey. 314 women (mean age 32.3 years) with dysmenorrhea reported a pain and cramps intensity of 7.45 ± 2.13 on a 0-10 Likert scale, which was reduced to 2.86 ± 1.81 upon treatment. Associated impairments of work/daily chores, leisure activities, and sleep were improved by 64.6%, 62.2% and 70.4%, respectively. The onset of symptom relief was within 60 min in 84.7%. Tolerability was rated as very good or good by 97.2%. These findings confirm the efficacy and tolerability data on PLUS from randomized controlled trials in a larger group of women conducting self-management of their dysmenorrhea.

Technological advancements must keep pace with earth’s rising demand for energy, while minimizing the carbon footprint on earth. One such option is using space based solar (SBS) radio frequency (RF) microwave power beaming. This innovative and novel technology is currently being explored to address terrestrial photovoltaic (PV) intermittency and provide a method to transmit power to other spacecraft in various orbits. This modality of power generation and distribution presents a new way to power terrestrial off-grid applications and earth orbiting spacecraft. In 1968, Dr. Peter Glaser published "Power from the Sun: Its Future", which serves as the foundation of this paper to revitalize scientific interest and efforts in this topic. This paper will primarily focus on the reliance of SBS initiatives on reusable launch vehicles (RLV) to place SBS spacecraft in a designated orbit, as well as the technological, economic, and operational challenges associated with power beaming to earth and other spacecraft. This includes a discussion of the variables from the operational environment that contribute or accelerate spacecraft end of life.

“Anti-stripping Agents" or "adhesion promoters” can enhance the chemical affinity between as-phalt and aggregate by increasing their mutual attraction. Various forms of anti-stripping agents have been proposed to mitigate pavement stripping, and siloxane-based Zychotherm is one of them. Choosing the appropriate type and dose of anti-stripping additives is no doubt vital to the intended performance. Therefore, it is critically important to determine the dose of the additives used in the modification of asphalt binders. This research developed a feasible detection method that can closely measure the dose (0.05% and 0.1%) of siloxane-based anti-stripping liquid agents. Related test methods including heat combustion test, residue visualization, burning, and ignition were implemented. Heat combustion results showed that with the addition of Zychotherm anti-stripping additive, average heat combustion value decreased by 1.34% and 1.72% for 0.05% and 0.1% Zychotherm modified binder respectively. In the burning and ignition process, the modified binder leaves yellowish substances in the residue, which is an indication of the presence of Zy-chotherm. The weight of the yellowish residue relates more to the quantity of Zychotherm in as-phalt binder.

We have successfully synthesized a series of Bi doped BaFe2As2 high quality single crystals for the first time. The x-ray diffraction (XRD) patterns show an expansion of lattice parameter c with Bi doping, indicating a negative pressure effect. By investigating the resistivity, a Fermi liquid (FL) to non-Fermi liquid (NFL) crossover is observed from normal state to antiferromagnetic order state, accompanied by three superconducting transitions labeled as SC I, SC II and SC III which are supposed to be induced from three superconducting realms with various Bi concentration. Thus, the NFL behavior is proposed to be closely related with the presence of superconductivity. The magnetic susceptibility measurements further speculates that the SC I and SC III phases should be filamentary superconductivity while the SC II be a possible bulk superconductivity with TC 7 K.

In his Cynefin classification of system behaviours, Snowden (2007), posits that his Simple (Clear ) and even Complicated system categories are sufficiently understandable to be amenable to the traditional methods of system analysis and modelling. Here, relationships between entities in the system are clear and unambiguous and they behave predictably. So, we can build mathematical “models” to describe how the system is expected to function in different situations. Leaving aside his fourth category, Chaotic, as too hard for the moment, this paper focusses on the third category, Complex systems. Yaneer Bar-Yam [1998] defined "complex systems" as systems that "have multiple interacting components, whose collective behaviour cannot be simply inferred from the behaviour of components." So, the traditional system modelling approach of decomposition into components and attempting to build up a picture of system performance from individual pieces, alone, is no longer appropriate for this class. The problem is that most of the methods we currently employ, and particularly the methods that attempt to make the case that these systems are safe, (such as Fault trees, FMEA. and Root Cause Analysis), rely on this decomposition into components and attempt to make the parts more reliable individually. Even the more systemic approaches, such as Reason’s (1990) Swiss Cheese Model, and Leveson’s System Theoretic Analysis (STPA) (2004), encourage adding layers of protection, barriers, strengthened control loops and better safety critical systems, in the hope that they are making the systems safer. But without a valid model to test the effectiveness of these add-ons, we cannot be sure that they will not have an opposite effect, of making the systems even more complex, unreliable and unpredictable. For complex systems, the careful gathering and documentation of system components and their relationships is still a necessary first step, but we need an additional capacity to model these complex systems to test out and prove they work safely and successfully in the real world. (not just draw fixed component relationships as system “models” (wiring diagrams)) In the last 10 years, an original approach developed by Hollnagel, (2012), to model systems as sets of interactive, interdependent “functions”, (abstracted from and independent of component details), has now been developed to the point where it can take the basic data and structures from the current component focussed system engineering “models”, such as MBSE( Goldberg (1994)), and can pull it all together in a dynamic system model from which analysts can discern how it really works in practice, and predict the emergent behaviours characteristic of complex systems. The paper describes how this methodology developed and how it is currently used to model real situations, to analyse incidents and anticipate events.

This review explores the field of retreatment strategies for the cases filled with calcium silicate-based root canal sealers. Since the introduction of calcium silicate-based materials in dentistry, calcium silicate-based root canal sealers have become popular among dentists because of their biocom- patibility, bioactivity, and sealing ability. Therefore, effective retreatment strategies are indispensable. This article aims to identify the challenges associated with the removal of calcium silicate-based sealers itself and gutta-percha with the sealers during retreatment, evaluate current techniques and materials, and provide future directions for research in this field. Regarding the strategies of removal of root canal sealers, calcium silicate-based sealers are still relatively new materials for clinicians compared with the traditional sealers such as epoxy- or eugenol-based sealers. First, no clinically established solvents have been reported. Second, calcium silicate-based sealers are currently utilized by clinicians in either cold sealer-based technique or warm vertical condensation technique. Third, the setting process of calcium silicate-based sealers generates byproducts primarily calcium hydroxide and secondarily hydroxyapatite, which could interact with dentine. Lastly, there is a lack of clinical studies evaluating the efficacy of retreatment protocols for teeth filled with calcium silicate-based sealers. Therefore, it is important to investigate chemo-mechanical properties of calcium silicate- based sealers itself and their reactions to solvents and/or mechanical instruments, and identify the interfacial properties of calcium silicate-based sealers in respect to dentine and gutta-percha. In addition, researchers in clinical field need to actively gather and report data on retreatments of teeth filled with calcium silicate-based sealers.

Road pavement monitoring represents the baseline to the pavement maintenance process. To quickly fix a damaged road, relevant authorities need a high-efficiency methodology that allows them to obtain data describing the current conditions of a network of roads. In urban areas, large-scale monitoring campaigns may be more expensive and not fast enough to describe how pavement degradation has evolved over time. Furthermore, at low speeds, many technologies are inadequate for monitoring the streets. In such a context, employing black box equipped vehicles to perform a routine inspection could be an excellent starting point. However, the vibration-based methodologies used to detect road anomalies are strongly affected by the speed of these vehicles. In this study, a statistical method was to analyze the effects of speed on road pavement conditions at different severity levels through the data recorded by taxi vehicles. Likewise introduced was a process that overcomes the speed effect when considering road surface conditions. The methodology proposed has succeeded in predicting the right damage severity level based on only two recorded parameters: speed and pavement deterioration index.

Rural areas of advanced industrial countries have frequently experienced a dualistic set of policies: one designed for the land-use sector, especially agriculture; and another dealing with rural development [1]. As problems of industrialised agriculture and forestry emerged, a more joined up approach to policy was advocated. Since the 1980s, at national and international level, new elements of rural policy have emerged to connect these two policy fields, including environmental payments, farm diversification grants, woodland creation within the farm sector and support for community-led local development. Rural land is also closely connected to the biodiversity and climate crises which have become important policy drivers. Policy towards community-led local development has moved further in Scotland than almost anywhere else as a result of policies for community-based land reform, community empowerment and community asset transfer. These communitarian approaches connect closely to collaborative landscape management as it addresses environmental challenges. As collaborative place-based and area-based approaches emerge as the guiding stars of the new rural development, this paper explores the challenges in the governance of sustainability transitions in moving from a sectoral rural policy hinging around support for land use to one based much more on communitarian values and place-based development.

The integration of robotics and image processing has led to the realization of robot autonomy in dynamic environments through the provision of visual feedback. This paper presents the application of parallel and open-link robots in palletizing and shape drawing tasks as enhanced by visual feedback from image processing. In determining the set of joint angles that could be used to reach the desired position and orientation of the end effector, the geometric approach in which the spatial geometry of the robotic arms was decomposed into several plane geometry problems was employed. Image processing techniques were used to enhance the performance of the robotic manipulators. In one approach, Color-based segmentation was used to distinguish between different objects in the workspace by using predefined color markers as references in the L*a*b color space. Classification of each pixel in the workspace image was then done by calculating the Euclidean distance between that pixel and a predefined color marker. A second approach employed Edge detection to identify the boundaries of objects within the workspace image by employing the Hough Transform mathematical model to detect the abrupt changes in the image brightness pixel-wise. The pixel locations from Hough were then sorted sequentially to outline the detected object. The integration of image processing with the robotic tasks was expected to improve the precise detection of the position of objects as well as the outline of geometric shapes. The incorporation of visual feedback allowed for dynamic robot manipulation in which prior knowledge of the workspace was not requisite. This led to improved pick and place as well as shape detection as applied in palletizing and shape drawing tasks actuated by the parallel and serial link manipulators, respectively.

In industry 4.0 era, video applications such as surveillance visual systems, video conferencing, or video broadcasting have been playing a vital role. In these applications, for manipulating and tracking objects in decoded video, the quality of decoded video should be consistent because it largely affects to the performance of the machine analysis. To cope with this problem, we propose a novel perceptual video coding (PVC) solution in which a full reference quality metric named Video Multimethod Assessment Fusion (VMAF) is employed together with a deep convolutional neural network (CNN) to obtain the consistent quality while still achieving the high compression performance. First of all, to achieve the consistent quality requirement, we propose a CNN model with an expected VMAF as input to adaptively adjust the quantization parameters (QP) for each coding block. Afterwards, to increase the compression performance, a Lagrange coefficient of Rate-Distortion optimization (RDO) mechanism is adaptively computed under Rate-QP and Quality-QP models. Experimental results show that the proposed PVC has achieved two targets simultaneously: the quality of video sequence is kept consistently with an expected quality level and the bit rate saving of the proposed method is higher than traditional video coding standards and relevant benchmark, notably with around 10% bitrate saving in average.

In recent years, the field of drug delivery has seen a significant shift towards the exploration and utilisation of nanoparticles (NPs) as versatile carriers for therapeutic agents. With its ability to provide exact control over NPs' characteristics, microfluidics has emerged as a potent platform for the efficient and controlled synthesis of NPs. Microfluidic devices designed for precise fluid ma-nipulation at the micro-scale offer a unique platform for tailoring NP properties, enabling en-hanced control over NP properties such as size, morphology and size distribution while ensuring high batch-to-batch reproducibility. Microfluidics can be used to produce liposomes, solid lipid nanoparticles, polymer-based NPs and lipid-polymer hybrid NPs, as well as a variety of inorganic NPs such as silica, metal, metal oxide, quantum dots and carbon-based NPs, offering precise control over composition and surface properties. Its unique precision in tailoring NP properties holds great promise for advancing NP-based drug delivery systems in both clinical and industrial set-tings. Although challenges with large-scale production still remain, microfluidics offers a trans-formative approach to NP synthesis. In this review, starting from the historical development of microfluidic systems, the materials used to create the systems, microfabrication methods and system components will be discussed in order to provide the reader with an overview of micro-fluidic systems. In the following, studies on the fabrication of nanoparticles such as lipid NPs, polymeric NPs and inorganic NPs in microfluidic devices are included.

Considering higher stability and lower cost of nanozymes than native enzymes, the nanozymatic systems had been utilized for several practical application, especially for sensing and detection. Most of common nanozymatic sensors are single-nanozyme based systems, however, recently a new generation of nanozyme-based systems called “multinanozyme system’ was introduced by Hormozi Jangi et al. (2020). Since the first report of multinanozyme systems, several multinaozyme systems have been developed and utilized for highly sensitive and selective sensing aims. The main advantages of multinaozyme systems compared of common nanozymatic sensors are their impact on simultaneous enhancing selectivity and sensitivity of sensor in a well-designed detection process. Since, the principles of design and detection mechanism of this new generation is not well-described in the literature, the aim of this article is the fast review of the principles of design of this new generation of nanozyme-based sensing and detection.

Massification in higher education has made the satisfaction of students’ needs a difficult conundrum among university lecturers. In contrast, the use of innovative design for collaborative learning enhances student engagement in the populous student’s context. Moreover, this paper aims to extensively explore innovative designs for fostering student engagement and collaborative learning among first-year students at the University of Venda. The study employed a qualitative research method with the purposive sampling technique. Subsequently, a group of 200 students was the population of this study. Participant Observation and narrative inquiry were used as data collection instruments. The students in their respective groups were assigned topics from the module content to use their talents to demonstrate their understanding of certain content in the module. The non-surprising findings of this paper elucidated that collaborative learning expedites students’ mastery of key concepts and subject content. The module lecturers introduced students to these innovative designs to ensure collaborative learning and effective student engagement. The key findings elucidated these aspects namely, role-playing exercises, group projects, peer-to-peer learning, use of talents, and peer feedback. The implication of this study is that students learn to work together, delegate responsibilities and communicate effectively to attain a common goal. Using these strategies, the lecturers promote collaboration among students and foster a more engaging and interactive learning experience. This paper further recommends efficient and effective techniques and strategies to foster student engagement and collaboration to track and monitor at-risk students timeously.

For women giving birth, every moment of delay in receiving skilled care significantly increases the risks of stillbirth, neonatal and maternal death. More than half of all births in developing countries including South Africa, take place outside a health facility and without skilled birth attendants. This has therefore, made it difficult to achieve the Sustainable Development Goals of global reduction in maternal mortality, which is a key health challenge globally, especially in developing countries and sub-Saharan Africa in particular. The aim of this study was to explore and describe the views of pregnant women regarding facility-based delivery. Focus group discussions were used to gather information from the pregnant women. Information was collected from six groups of pregnant women who had delivered babies at the antenatal care facilities in the past years. Results showed several factors associated with the failure to use institutional delivery service, such as long distance from the health care facility, lack of transport, lack of transport fare, shortages of skilled staff, failure to disclose pregnancy, cultural and religious beliefs, and staff attitudes.

Model-based approaches form the basis of the majority of damage detection and localization studies; however, built-in online detection technique can offer an alternative for future structural health monitoring technologies. Thus, in this paper, we present a dynamic localization algorithm for damage localization in functionally graded plates. The method depends on the creation of a grid matrix that contains the dynamic response of the structure over time. Subsequently, an optimization process is carried out via a linear equation representing the information contained within the grid to achieve accurate localization of damage. Due to the inherent sparsity of the localization nature, the FISTA algorithm has been employed as a problem solver. The approach is tested in the case of the functionally graded plate under clamped free boundary conditions. Several, damage scenarios were investigated for the case of damage signals on and off the grid. The results show that the proposed approach is capable of accurately predicting damage position and can be suitable for use in low size data systems.

Diabetes mellitus (DM) comprises a range of metabolic disorders characterized by high blood glucose levels caused by defects in insulin release, insulin action, or both. DM is a widespread condition that affects a substantial portion of the global population, causing high morbidity and mortality rates. The prevalence of this major public health crisis is predicted to increase in the forthcoming years. Although, several drugs are available to manage DM, these are associated with adverse side effects, which limits their use. In underdeveloped countries, where such drugs are often costly and not widely available, many people continue to rely on alternative traditional medicine, including medicinal plants. The latter serve as a source of primary healthcare and plant-based foods in many low and middle-income countries. Interestingly, many of the phytochemicals they contain have been demonstrated to possess antidiabetic activity such as lowering blood glucose levels, stimulating insulin secretion and alleviating diabetic complications. Therefore, such plants may provide protective effects that could be used in the management of DM. The purpose of this article was to review the medicinal plant-based foods traditionally used for the management of DM, including their therapeutic effects, pharmacologically-active phytoconstituents and antidiabetic mode of action at the molecular level. It also presents future avenues for research in this field.