Single-cell RNA sequencing (scRNA-seq) technology has significantly advanced our understanding of the diversity of cells and how this diversity is implicated in diseases. Yet, translating these findings across various scRNA-seq datasets poses challenges due to technical variability and dataset-specific biases. To overcome this, we present a novel approach that employs both an LLM-based framework and explainable machine learning to facilitate generalization across single-cell datasets and identify gene signatures to capture disease-driven transcriptional changes. Our approach uses scBERT, which harnesses shared transcriptomic features among cell types to establish consistent cell-type annotations across multiple scRNA-seq datasets. Additionally, we employ a symbolic regression algorithm to pinpoint highly relevant yet minimally redundant models and features for inferring a cell type’s disease state based on its transcriptomic profile. We ascertain the versatility of these cell-specific gene signatures across datasets, showcasing their resilience as molecular markers to pinpoint and characterize disease-associated cell types. Validation is carried out using four publicly available scRNA-seq datasets from both healthy individuals and those suffering from ulcerative colitis (UC). This demonstrates our approach’s efficacy in bridging disparities specific to different datasets, fostering comparative analyses. Notably, the simplicity and symbolic nature of the retrieved gene signatures facilitate their interpretability, allowing us to elucidate underlying molecular disease mechanisms using these models.

With the advancement of urbanization in China, effective control of pollutant emissions and air quality have become important goals in current environmental management. Nitrogen dioxide (NO2), as a precursor of tropospheric ozone and fine particulate matter, plays a significant role in atmospheric chemistry research and air pollution control. However, the uneven ground monitoring stations and low temporal resolution of polar-orbiting satellites set challenges for accurately assessing near-surface NO2. To address this issue, a spatiotemporal refined NO2 retrieval model was established for China using the geostationary satellite Himawari-8. The spatiotemporal characteristics of NO2 were analyzed and its contribution factors were explored. Firstly, seven Himawari-8 channels sensitive to NO2 were selected by using the forward feature selection based on information entropy. Subsequently, a 2DCNN-LSTM network model was constructed, incorporating the selected channels and meteorological variables as retrieval factors to estimate hourly NO2 in China from March 2018 to February 2020 (with a resolution of 0.05°, per hour). The performance evaluation demonstrated that the full-channel 2DCNN-LSTM model had good fitting capability and robustness (R2=0.74, RMSE=10.93), and further improvements were achieved after channel selection (R2=0.87, RMSE=6.84). The 10-fold cross-validation results indicated that the R2 between retrieval and measured values was above 0.85, the MAE was within 5.60, and the RMSE was within 7.90. R2 varied between 0.85 and 0.90, showing better validation at mid-day (R2=0.89) and in spring and fall transition seasons (R2 =0.88 and R2 =0.90). To investigate the cooperative effect of meteorological factors and other air pollutants on NO2, statistical methods (Beta coefficients) were used to test the factor interpretability. Meteorological factors as well as other pollutants were analyzed. From a statistical perspective, PM2.5, Boundary Layer Height, and O3 were found to have the largest impacts on near-surface NO2, with each standard deviation change in these factors leading to 0.28, 0.24, and 0.23 in standard deviations of near-surface NO2, respectively. Findings of the study contribute to a comprehensive understanding of the spatiotemporal distribution of NO2 and provide a scientific basis for formulating targeted air pollution policies.

Software Defined Networking (SDN) is positioning the standard for the management of networks due to its scalability and flexibility to program the network. The SDN provides many advantages but it also involves some specific security problems take down the controller using cyber attack and in result the whole network will shut down which makes it a single point of failure. In this paper, the DDoS attacks in SDN were detected using AI-enabled, machine and deep learning, models with some specific features for data-set under normal and DDoS traffic. In our approach, initial data-set is collected from 84 features on kaggle and then the 20 top most features are selected using permutation importance algorithm. The data-set were learned and tested with AI-enabled 5 models. Our experimental results showed that the use of machine learning based random forest model has achieved the highest accuracy rate of 99.97%, in DDoS attack detection in SDN. Our contributions through this study are, first, we found highest 20 attacks that absolutely contributed to DDoS attacks. Secondly, it can reduce the time and cost of comparing various learning models and performance required for determining a learning model suitable for DDoS detection. Finally, various experimental methods for evaluating the performance of the learning model are presented so that related researchers can utilize them.

Electroencephalography (EEG) signals provide valuable insights into various activities of the human brain, including the detection of mental stress, which is a complex physiological and psychological response. However, the challenge lies in identifying mental stress accurately while mitigating the limitations associated with a large number of EEG channels. This includes issues such as computational complexity, the risk of overfitting, and the increased setup time for electrode placement, which can be cumbersome for real-life applications. Therefore, it is crucial to develop EEG channel selection algorithms that enable the creation of a wearable device capable of assessing mental stress in real-life scenarios. This study introduces a novel channel selection method aimed at identifying highly accurate channels for detecting mental stress. Our approach, known as the Correlation Coefficient of Hjorth Parameters (CCHP), assesses the correlation between activity, mobility, and complexity in the time domain to nominate the most relevant channels. By selecting channels that exhibit high correlation with the stress task while being uncorrelated with each other, CCHP significantly reduces the number of EEG channels required, without compromising accuracy or performance. To evaluate the effectiveness of CCHP, we conducted experiments using the DEAP public dataset. Comparing our results with other recent algorithms that utilize the full set of EEG channels, CCHP achieved a superior classification accuracy of 81.56% using only eight EEG channels. Furthermore, CCHP outperformed existing channel selection methods by an impressive 8%. These findings strongly indicate that the CCHP algorithm shows great promise in the design of a wearable application for mental stress detection, utilizing a minimal number of EEG channels.

In this paper we derive a unified perspective for Optimal Transport (OT) theory and Mean Field Control (MFC) theory to analyse the learning process for Neural Networks algorithms in a high-dimensional framework. We consider Mean Field Neural Networks in the context of MFC theory, specifically the mean field formulation of OT theory that allows the development of highly efficient algorithms while providing a powerful tool in the context of explainable Artificial Intelligence.

Like the most of crops, especially sugarcane needs to be kept upright until it is harvested. Because, lodging of sugarcane has a significant negative effect on cane yield and sugar content of sugarcane. To keep sugarcane upright earthing up is an utmost intercultural operation. In Bangladesh, most of the operations for sugarcane cultivation including earthing up are generally done in traditional method using human labor which increases production costs as well as reduces the income of sugarcane growers. So, a cost effective two-wheeled tractor (2WT) mounted earthing up machine was developed in Bangladesh Sugarcrop Research Institute (BSRI), Pabna to reduce drudgery and cost of sugarcane production. Field tests were conducted in experimental sugarcane field at BSRI and technical and economic performances of the developed earthing up machine were also carried out based on the field test. Average effective field capacity and field efficiency were found 0.16 ha/hr and 77.41 %, respectively for the developed earthing up machine. The earthing up machine was not found economically viable when it is used only for earthing up operation. Besides, when the 2WT was being used as the main driver for other activities including earthing up operation, then the earthing up machine becomes economically beneficial with net cash flow, net present value, internal rate of return, benefit-cost ratio and payback period of BDT 148497/ha, BDT 23184, 3%, 3.81:1 and approximately 1 year, respectively. On the contrary, considering cost of only earthing up tool without 2WT, then it was found economically beneficial with net cash flow, internal rate of return, net present value, benefit-cost ratio and payback period of BDT 16428/ha, BDT 3053, 4.7%, 2.71:1 and approximately 2 years, respectively. Since 2WT is commonly used for versatile farming purposes. Therefore, versatile use of 2WT as prime mover for other machines including earthing up machine can make earthing up machine economically viable and beneficial for sugarcane growers.

With the rise of piano teaching in recent years, many people have joined the team of piano learners. However, the expensive cost of manual instruction and the unique one-on-one teaching model have made piano learning an extravagant event. Most existing approaches based on the audio modality aim to evaluate piano players' skills. Unfortunately, these methods ignored the information contained in the video, which led to a one-sided and simplistic evaluation of the piano player's skills. More recently, multimodal-based methods are proposed to assess the skill level of piano players using both video and audio information. However, existing multimodal approaches use shallow networks to extract video and audio features, which are deficient in extracting complex spatio-temporal and time-frequency features from piano performance. Furthermore, the fingering and the pitch-rhythm information of the piano performance is contained in the spatio-temporal and time-frequency features, respectively. In the paper, we propose a ResNet-based audio-visual fusion model that combines video and audio features to assess the skill level of piano players. Firstly, ResNet18-3D is used as the backbone network for our visual branches, which can extract feature information from the video data. Then, we consider ResNet18-2D as the backbone network of the aural branch and extract the feature information from the audio data. The extracted video features are fused with the audio features to generate multimodal features for the final piano skill evaluation. The experimental results on the PISA dataset show that our proposed audio-visual fusion model, with a validation accuracy of 70.80%, outperforms the state-of-the-art methods in both performance and efficiency. Then, we also explore the impact of different layers of ResNet on model performance, and the experimental results show that the audio-visual fusion model dealing with the piano skill assessment problem can make full use of both feature information when the number of video features is close to the number of audio features.

Data scarcity is a significant challenge in the field of Machine Learning (ML), as data collection can be expensive, time-consuming, and difficult, particularly in developing countries. This challenge is exaggerated on the need to use dataset for livestock disease predictions for early intervention and surveillance. To address this challenge, this paper presents a data synthesis method that has been used to accurately generate new data samples from few real-world data. With much data available to train the ML models, overfitting is eliminated. We present the use of Generative Adversarial Networks mainly the Conditional Tabular Generative Adversarial Network to synthesize categorical data for training machine learning models for prediction of the Pestes des Petits Ruminants (PPR) disease. The results showed that training score became 0.89 and the cross-validation score was 0.87 after synthesized data was used with Random Forest algorithm. The resulting dataset can be used to support the prediction and surveillance of the Pestes des Petits Ruminants (PPR) disease. The proposed method can also be applied to any domain with categorical data, and has the potential to improve the performance of machine learning models with increased data availability.

Conducting related research is crucial since there are still many problems that need to be resolved in the research on soft robots in the areas of material selection, structure design and manufacture, and drive control. Soft manipulators, a subset of soft robots, are now a popular area of study for many researchers. In comparison to typical manipulators, soft manipulators feature a high degree of gripping flexibility and a basic morphological structure. They are composed of flexible materials. It has a wide range of potential applications in healthcare, rehabilitation, bionics, and detection, and it can compensate for the drawbacks of rigid manipulators in some use scenarios. A modular soft-body torsional gripping system is developed after a torsional and gripping actuator is conceived, constructed, and its performance is examined. The torsion actuator and the grasping actuator can be combined in the system in a modular fashion. With the help of RGB-D vision algorithms, this multi-modular setup makes it possible to combine soft actuators with various twisting degrees and achieve exact gripping. Through pneumatic control, the target object is precisely grasped and rotated at various angles, enabling rotation of the target object in three dimensions.

This paper mainly used database technology, machine learning, thermodynamic calculation, experimental verification, etc., on integrated computational materials engineering. The interaction between different alloying elements and the strengthening effect of precipitated phases are investigated mainly for martensitic ageing steels. Modelling and parameter optimization were performed by machine learning, and the highest prediction accuracy was 98.58%. Investigated the influence of composition fluctuation on performance and correlation tests to analyze the influence of elements from multiple perspectives. Furthermore, screen out the three-component composition process parameters with composition and performance with high contrast. Thermodynamic calculations studied the effect of alloying element content on the nano-precipitation phase, Laves phase, and austenite in the material. The heat treatment process parameters of the new steel grade were also developed based on the phase diagram. A new type of martensitic ageing steel was prepared by selected vacuum arc melting. The sample with the highest overall mechanical properties had a yield strength of 1887 MPa, a tensile strength of 1907 MPa, and a hardness of 58 HRC. The sample with the highest plasticity had an elongation of 7.8%. The machine learning process for the accelerated design of new ultra-high tensile steels was found to be generalizable and reliable.

Abstract: Despite existing several techniques for distributed sensing (temperature and strain) using standard Single Mode optical Fiber (SMF), compensating or decoupling both effects is mandatory for many applications. Currently, most of the decoupling techniques require special optical fibers and are difficult to implement with high spatial resolution distributed techniques, such as ϕ-PA-OFDR. So, this work’s objective is to study the feasibility of decoupling temperature and strain out of a ϕ-PA-OFDR readouts taken over an SMF. For this purpose, the readouts will be subjected to a study using several Machine Learning algorithms, among them, Deep Neural Networks. The motivation which underlies this target is the current blockage in the widespread use of Fiber Optic Sensors in situations where both strain and temperature change, due to the coupled dependence of currently developed sensing methods. Instead of using other types of sensors or even other interrogation methods, the objective of this work is to analyze the available information in order to develop a sensing method capable of providing information about strain and temperature simultaneously.

In the last decade, the genetic engineering world has been shaken up by a relatively new genetic editing tool based on RNA-guided Nucleases (RGNs): the CRISPR/Cas9 system. Since the first report in 1987 and its characterization in 2007 as a bacterial defense mechanism, the interest and research on this system have grown exponentially. CRISPR systems provide immunity to bacteria against invading genetic material; however, with specific modifications in sequence and structure, it becomes a precise editing system that makes it possible to genetically modify almost any organism. There are diverse approaches regarding the refinement of these modifications, such as constructing more accurate nucleases, understanding the cellular context and facing the epigenetic conditions, or re-designing guide RNAs (gRNAs). Considering the critical importance for the correct CRISPR/Cas9 systems performance, our scope will emphasize in the latter approach. Hence, we present an overview of the past and the most recent guide RNA web-based design tools, highlighting their computational architecture and gRNA characteristics evolution through the years. Our study concisely explains the computational approaches that use machine learning techniques, deep neural networks, and large datasets of gRNA/target interactions to make possible both predictions and classifications directed to design, optimize, and create promising gRNAs suitable for future gene therapies.

The current epoch, known as the Anthropocene, is characterized by significant human impact on the global ecological system, including altered species survival and distribution patterns due to global warming resulting from industrialization. Accurately quantifying and assessing the extent of human impact on the environment is of utmost importance. To this end, four primary methods have been developed to measure human impact: (1) ecological footprint, (2) Human appropriation of net primary production (HANPP), (3) Planetary boundary, and (4) Living Planet Index. In this paper, we aim to provide a comprehensive understanding of these methods, including their strengths and weaknesses, to evaluate their efficacy in accurately assessing the impact of human activities on the global ecosystem. Furthermore, we explore the potential of utilizing novel machine learning modelling for future predictions in this area. By analyzing and comparing these methods, we can gain insight into the ways in which human activities impact the environment and use this knowledge to inform future policies and practices to mitigate negative effects.

Backgrounds: The field of Artificial Intelligence (AI) has seen a major shift in recent years due to the development of new Machine Learning (ML) models such as Generative Pre-trained Transformer (GPT) and AI which are progressively becoming part our everyday lives. GPT has achieved previously unheard-of levels of accuracy in most computerized language processing tasks and their chat-based variations. Aim: The aim of this study was to investigate the problem-solving abilities of ChatGPT using two sets of verbal insight problems, with a known performance level established by a sample of human participants. Material and Methods: A total of 30 problems labelled as “practice problems” and “transfer problems”, as listed by Ansburg and Dominowski (2000), were administered to ChatGPT. The answers provided by ChatGPT received a score of "0" for each problem answered incorrectly and a score of "1" for each correct response, as per the correct solution specified in Ansburg and Dominowski's (2000) study. The highest score that could be attributed to both the practice and transfer problems was 15 out of 15. In order to compare ChatGPT performance to the performance to that of human subjects, the solution rate for each problem (based on a sample of 20 subjects) was used, as indicated by Ansburg and Dominowski (2000). Results: The study highlighted that ChatGPT can be trained in out-of-the-box thinking and demonstrated potential in solving verbal insight problems. The global performance of ChatGPT equalled the most probable outcome for the human sample in both practice problems and transfer problems as well as upon their combination. Additionally, ChatGPT answer combinations were among the 5% of most probable outcomes for the human sample both when considering practice problems and pooled problem sets. These findings demonstrate that ChatGPT performance on both set of problems was in line with the mean rate of success of human subjects, indicating that it performed reasonably well. Conclusions: The use of transformer architecture and self-attention in ChatGPT may have helped to prioritize inputs while predicting, contributing to its potential in verbal insight problem-solving. ChatGPT has shown potential in solving insight problems, thus highlighting the importance of incorporating AI in psychological research. However, it is acknowledged that open challenges still exist. Indeed, further research is needed to fully understand the capabilities and limitations of AI in insight problem-solving.

Reliable and accurate flood prediction is a challenging task in poorly gauged basins due to data scarcity. Data is an essential component of any AI/ML model today, and the performance of such models hugely depends on the availability of sufficient amount of trusted, representative data. However, unlike a few well-studied rivers, most of the rivers in developing countries are still insufficiently monitored, which significantly hinges the design and development of advanced flood prediction models and early warning systems. This paper presents a multi-modal, sensor-based and near-real time river monitoring system to produce a multi-feature data set for the Kikuletwa river in Northern Tanzania, an area that heavily suffers from frequent floods. Our deployed system, which gather information about river depth levels and weather at several locations, aims at widening the ground truth of the river characteristics and eventually improve the accuracy of flood predictions. We provide details on the monitoring system used to gather the data as well as report on the methodology and the nature of the data. Finally, we present the relevance of the data set in the context of flood prediction, discussing the most suitable AI/ML-based forecasting approaches, while also highlighting some applications of the data set beyond flood warning systems.

Ovarian cancers are curable by surgical resection when discovered early enough. Unfortunately, most ovarian cancers are diagnosed in the later stages. One strategy to identify early ovarian tumors is to screen women who have the highest risk scores. This mini review summarizes the accuracy of different methods used to assess the risk of developing ovarian cancer, including family history, BRCA genetic tests, and polygenic risk scores. The accuracy of these is compared to the maximum theoretical accuracy, revealing a substantial gap. We suggest that this gap, or missing heritability, could be caused by epistatic interactions between genes. An alternative approach to computing genetic risk scores, using chromosomal-scale length variation should incorporate epistatic interactions. Future research in this area should focus on this and other alternative methods of characterizing genomes.

Background: There is an increasing interest in the role of EEG in neurorehabilitation. We primarily aimed to identify the knowledge base through highly influential studies. Our secondary aims were to imprint the relevant thematic hotspots, research trends, and social networks within the scientific community. Methods: We performed an electronic search in Scopus looking for studies reporting on rehabilitation in patients with neurological disabilities. The most influential papers outlined the knowledge base, while a word co-occurrence analysis imprinted the research hotspots. Likewise, co-citation analyses highlighted collaboration networks between Universities, authors, and countries. The results were presented in summary tables, burst detection plots, and geospatial maps. Finally, a content review based on the top-20 most cited articles completed our study. Results: Our current bibliometric study was based on 874 records from 420 sources. There was a vivid research interest in EEG use for neurorehabilitation, with an annual growth rate as high as 14.3%. The most influential paper was the study titled "Brain-computer interfaces, a review" by Nicolas-Alfonso LF and Gomez-Gill J, with 997 citations, followed by "Brain-computer interfaces in neurological rehabilitation" by Daly J. and Wolpaw JR (708 citations). The USA, Italy, and Germany were among the most productive countries. The research hotspots shifted with time from the use of “functional magnetic imaging” to EEG-based “brain-machine interface”, “motor imagery”, and “deep learning”. Conclusions: EEG constitutes the most significant input in brain-computer interfaces (BCI) and can be successfully used in the neurorehabilitation of patients with stroke, amyotrophic lateral sclerosis, and traumatic brain and spinal injury. EEG-based BCI facilitates training, communication, and control of wheelchair and exoskeletons. However, research is limited to specific scientific groups from developed countries. Evidence is expected to change with the broader availability of BCI and improvement in EEG filtering algorithms.

The evolving global SARS-CoV-2 pandemic emphasizes how unprepared we are for the emergence of the next lethal viral pathogen. A list of potential candidates was created by the World Health Organization which named Nipah virus infection as the highly lethal prototypic member of that list. Building on our earlier viral pandemic preparedness research into SARS-CoV-2 we have created computer simulations of Nipah virus infection complicated by encephalitis, the most common cause of Nipah-associated mortality. In the current experiments, we first created updated simulations of wild-type whole-brain organoids (aiWBO). Upon validation, the aiWBO were infected with the simulated Nipah virus genome. The Nipah encephalitis simulations (aiWBO-NiV) were then used to find optimal single, double, and triple protein combinations for candidates as potential targets for vaccine development. Our data suggest that the use of multi-viral proteins/epitopes chimera is the most promising approach to Nipah vaccine development and that employing artificial intelligence to guide the identification of promising vaccine candidates is an efficient and cost-effective strategy for future viral pandemic preparedness.

According to the World Health Organization, about 15% of the world’s population has some form of disability. Assistive Technology, in this context, contributes directly to the overcoming of difficulties encountered by people with disabilities in their daily lives, allowing them to receive education and become part of the labor market and society in a worthy manner. Assistive Technology has made great advances in its integration with Artificial Intelligence of Things (AIoT) devices. AIoT processes and analyzes the large amount of data generated by IoT devices and applies Artificial Intelligence models, specifically Machine Learning, to discover patterns for generating insights and assisting in decision making. Based on a systematic literature review, this article aims at identifying the Machine Learning models used in multiple different research about Artificial Intelligence of Things applied to Assistive Technology. The survey of the topics approached in this article also highlights the context of such research, their application, IoT devices used, and gaps and opportunities for further development. Survey results show that 50% of the analyzed research address visual impairment, and for this reason, most of the topics cover issues related to computational vision. Portable devices, wearables, and smartphones constituted the majority of IoT devices. Deep Neural Networks represent 81% of the Machine Learning models applied in the reviewed research.

Heme proteins serve diverse and pivotal biological functions. Therefore, clarifying the mechanisms of these diverse functions of heme is a crucial scientific topic. Distortion of heme porphyrin is one of the key factors regulating the chemical properties of heme. Here, we constructed convolutional neural network models for predicting heme distortion from the tertiary structure of the heme-binding pocket to examine their correlation. For saddling, ruffling, doming, and waving distortions, the experimental structure and predicted values were closely correlated. Furthermore, we assessed the correlation between the cavity shape and molecular structure of heme and demonstrated that hemes in protein pockets with similar structures exhibit near-identical structures, indicating the regulation of heme distortion through the protein environment. These findings indicate that the tertiary structure of the heme-binding pocket regulates the distortion of heme porphyrin, thereby controlling the chemical properties of heme relevant to the protein function; this implies a structure–function correlation in heme proteins.

Gestational diabetes mellitus (GDM) is a common complication of pregnancy that adversely affects maternal and offspring health. A variety of risk factors, such as BMI and age, have been associated with increased risks of gestational diabetes. However, in many cases gestational diabetes occurs in healthy nulliparous women with no obvious risk factors. Emerging data suggest that the tendency to develop gestational diabetes has genetic and environmental components. Here we develop a polygenic risk score for gestational diabetes. We further investigate relationships between the genetic architecture of GDM and genetically constructed risk factors and biomarkers. Our results show that genetics can be used as an early screening tool that identifies at-risk at higher risk of GDM before its onset to propose comprehensive monitoring and preventative programs to mitigate the risks.

The aim of this tutorial is to help students grasp the theory and applicability of support vector machines (SVMs). The contribution is an intuitive style tutorial that helped students gain insights into SVM from a unique perspective. An internet search will reveal many videos and articles on SVM, but many of them give simplified explanations that leave gaps in the derivations that beginning students cannot fill. Most free tutorials lack guidance on practical applications and considerations. The software wrappers in popular programming libraries such as Python and R hide many of the operational complexities. Free software tools often use default parameters that ignore domain knowledge or leave knowledge gaps about the important effects of SVM hyperparameters, resulting in misuse and subpar outcomes. The author uses this tutorial as a course reference for students studying artificial intelligence and machine learning. The tutorial derives the classic SVM classifier from first principles and then derives the practical form that a computer uses to train a classification model. An intuitive explanation about confusion matrices, F1 score, and the AUC metric extend insights into the inherent tradeoff between sensitivity and specificity. A discussion about cross-validation provides a basic understanding of hyperparameter tuning to maximize generalization by balancing underfitting and overfitting. Even seasoned self-learners with advanced statistical backgrounds have gained insights from this tutorial style of intuitive explanations, with all related considerations for tuning and performance evaluations in one place.

In the present study, we investigate the post-editese phenomenon, i.e., the unique features that set machine translated post-edited texts apart from human-translated texts. We use two literary texts, namely, the English children's novel by Lewis Carroll Alice’s Adventures in Wonderland (AW) and Paula Hawkins' popular book The Girl on the Train (TGOTT) translated from English into Brazilian-Portuguese to investigate whether the post-editese features can be found on the surface of the post-edited (PE) texts. In addition, we examine how the features found in the PE texts differ from the features encountered in the human-translated (HT) and machine translation (MT) versions of the same source text. Results revealed evidence for post-editese for TGOTT only with PE versions being more similar to the MT output than to the HT texts.

Communication is indispensable for today's lifestyle, and thanks to technology, millions of people can communicate as quickly as possible. The effect of this breakthrough has transformed organizations to the degree that they generate billions of emails daily to facilitate their operations. There is implicit information behind this vast corpus of human-generated content that can be mined and used for their benefit. This paper tries to address the opportunity that email logs can bring to organizations and propose an approach to discover process models by combining supervised text classification and process mining. This framework consists of two main steps, text classification, and process mining. First, Emails will be classified with supervised machine learning, and to mine, the processes fuzzy Miner is used. To further investigate the application of this framework, we also applied this framework over a real-life dataset from a case study organization.

Background: Gastric cancer (GC) is one of the leading malignancy diseases worldwide, especially in Asian. CAST is a potential oncogene in GC carcinogenesis process. The character of macrophage infiltration in GC microenvironment was also unaddressed. Methods: We first applied machine searching in gene candidate evaluation of GC. CAST expression was analyzed via the Human Protein Atlas (HPA) and Gene Expression Profiling Interactive Analysis 2 (GEPIA2) database. Protein-protein interaction (PPI) network was downloaded from STRING. We investigated the impact of CAST on clinical prognosis using Kaplan-Meier plotter. The correlations between CAST and Lgr5 and macrophage infiltration in GC was surveyed via TIMER 2.0. Finally, GeneMANIA was also used to evaluate the possible functional linkage between genes. Results: After machine-assisted searching, CAST expression was found signicant difference in the overall survival of GC patients. STRING revealed CAST related proteomics and transcriptomics associations, mainly about CAPN family. Moreover, CAST significantly impacts the prognosis of GC from other datasets validation. Notably, high CAST expression was correlated with worse overall survival in GC patients (hazard ratio = 1.59; logrank P = 9.4 x 10-8). CAST and Lgr5 expressions were both positively correlated with WNT 2 and WNT 2B. Among GC patients in several datasets, CAST and macrophage infiltration evaluated together showed no obvious trend toward poor clinical overall survival. Conclusion: CAST plays an important role in GC clinical prognosis and is associated with WNT 2/WNT 2B/Lgr5. Our study denmostrated that CAST in GC overall survival is regulated by macrophage infiltration.

Stumbling during gait is commonly encountered in patients who suffer from mild to serious walking problems, e.g. after stroke, in osteoarthritis, or amputees using a lower leg prosthesis. Instead of self-reporting, an objective assessment of the amount of stumbles in daily life would inform clinicians more accurately and enable the evaluation of treatments that aim to achive a safer walking pattern. An easy to use wearable might fullfill this need. The goal of the present study was to investigate whether a single inertial measurement unit (IMU) placed at the shank and machine learning algorithms could be used to detect and classify stumbling events in a dataset comprising of a wide variety of daily movements. Ten healthy test subjects were deliberately tripped by an unexpected and unseen obstacle while walking on a treadmill. The subjects stumbled a total of 276 times, both using an elevating recovery strategy and a lowering recovery strategy. Subjects also performed multiple Activities of Daily Living. During data processing, an event-defined window segmentation technique was used to trace high peaks in acceleration which could potentially be stumbles. In the reduced dataset, time windows were labelled with the aid of video annotation. Subsequently, discriminative features were extracted and fed to train seven different types of machine learning algorithms. Trained machine learning algorithms were validated using leave-one-subject-out cross-validation. Support Vector Machine (SVM) algorithms were most succesful, and could detect and classify stumbles with 100% sensitivity, 100% specificity and, 96.7% accuracy, in the independent testing dataset. The SVM algorithms were implemented in a user-friendly, freely available, stumble detection app named Stumblemeter. This work shows that stumble detection and classification based on SVMs is accurate and ready to apply in clinical practise.

Mental stress is one of the serious factors that lead to many health problems. Scientists and physicians have developed various tools to assess the level of mental stress in its early stages. Several neuroimaging tools have been proposed in the literature to assess mental stress in the workplace. Electroencephalogram (EEG) signal is one important candidate because it contain rich information about mental states and condition. In this paper, we review the existing EEG signal analysis methods on the assessment of mental stress. The review highlights the critical differences between the research findings and argues that variations of the data analysis methods contribute to several contradictory results. The variations in results could be due to various factors including lack of standardized protocol, the brain region of interest, stressor type, experiment duration, proper EEG processing, feature extraction mechanism, and type of classifier. Therefore, the significant part related to mental stress recognition is choosing the most appropriate features. In particular, a complex and diverse range of EEG features, including time-varying, functional, and dynamic brain connections, requires integration of various methods to understand their associations with mental stress. Over this, the review suggests fusing the cortical activations with the connectivity network measures and deep learning approaches to improve the accuracy of mental stress level assessment.

As air traffic demand grows, robust, data-driven anomaly detection methods are required to ensure that aviation systems become safer and more efficient. The terminal airspace is identified as the most critical airspace for both individual flight-level and system-level safety and efficiency. As such, developing data-driven anomaly detection methods to analyze terminal airspace operations is paramount. With the expansion of ADS-B technology, open-source flight tracking data has become more readily available to enable larger-scale analyses of aircraft operations. This paper makes a distinction between spatial metrics in ADS-B trajectory data and energy metrics derived from ADS-B trajectory data. Motivated by the limited number of approaches that simultaneously consider both spatial and energy metrics, this paper introduces the concepts of spatial anomalies and energy anomalies. In particular, it proposes a novel, unified framework for detection of spatial and energy anomalies in ADS-B trajectory data (and associated derived metrics). The framework consists of three main parts - a data processing procedure, a spatial anomaly detection method, and an energy anomaly detection method. The framework is demonstrated utilizing four months of ADS-B trajectory data associated with arrivals at San Francisco International Airport, and the relationship between the spatial and energy anomalies in this terminal airspace is explored. The results that stem from the implementation of this framework indicate that if an aircraft is spatially not conforming to an identified set of air traffic flows representing standard spatial operations, then this aircraft is more likely to experience non-conformance to standard operations in its energy metrics. Aviation operators, such as air traffic controllers, may benefit from this observation, as it may factor into decision-making in instances where there is the potential to instruct an aircraft to spatially deviate from standard operations. Additionally, this research revealed underlying differences between trajectories that are spatially nominal yet energy-anomalous and those trajectories that are spatially anomalous and energy-anomalous. Focusing solely on energy anomaly detection does not provide insight into potential spatial-related decisions that may have been made to result in off-nominal energy behavior.

Conventionally, partial differential equations (PDE) problems are solved numerically through discretization process by using finite difference approximations. The algebraic systems generated by this process are then finalized by using an iterative method. Recently, scientists invented a short cut approach, without discretization process, to solve the PDE problems, namely by using machine learning (ML). This is potential to make scientific machine learning as a new sub-field of research. Thus, given the interest in developing ML for solving PDEs, it makes an abundance of an easy-to-use methods that allows researchers to quickly set up and solve problems. In this review paper, we discussed at least three methods for solving high dimensional of PDEs, namely PyDEns, NeuroDiffEq, and Nangs, which are all based on artificial neural networks (ANNs). ANN is one of the methods under ML which proven to be a universal estimator function. Comparison of numerical results presented in solving the classical PDEs such as heat, wave, and Poisson equations, to look at the accuracy and efficiency of the methods. The results showed that the NeuroDiffEq and Nangs algorithms performed better to solve higher dimensional of PDEs than the PyDEns.

The color and NIR spectrum are key to build an oil estimation model, thus it requires individual olives clustering before the Sohlext oil extraction method can be applied. The objective was to analyze an OC estimation model of individual olives, based on cluster of similar color and NIR spectrum in different combination of the first and/or the second season. This study was performed with Chilean Arbequina olives in 2016 and 2017. The descriptor of the cluster consisted of the 3 color channels of c1, c2, c3 color model plus 11 reflectance points between 1710 and 1735 nm of each olive, normalized with the Z-score index. Clusters of similar color and NIR spectrum were formed with the k-means++ algorithm, leaving a sufficient amount of olives to be able to perform the Sohlext analysis of OC, as reference value. The estimation models were based on the Support Vector Machine. The test was carried out with the Leave One-Out Cross Validation in different training-testing combinations. The best model predicted the OC with 6% and 13%deviation respect to the real value in one season by itself and when one season tested with another season, respectively. The use of clustering in estimation model is discussed.

The adoption of Machine Learning (ML) for building emulators for complex physical processes has seen an exponential rise in the recent years. While neural networks are good function approximators, optimizing the hyper-parameters of the network to reach a global minimum is not trivial, and often needs human knowl- edge and expertise. In this light, automatic ML or autoML methods have gained large interest as they automate the process of network hyper-parameter tuning. In addition, Neural Architecture Search (NAS) has shown promising outcomes for improving model performance. While autoML methods have grown in popularity for image, text and other applications, their effectiveness for high-dimensional, complex scientific datasets remains to be investigated. In this work, a data driven emulator for turbulence closure terms in the context of Large Eddy Simulation (LES) models is trained using Artificial Neural Networks and an autoML frame- work based on Bayesian Optimization, incorporating priors to jointly optimize the hyper-parameters as well as conduct a full neural network architecture search to converge to a global minima, is proposed. Additionally, we compare the effect of using different network weight initializations and optimizers such as ADAM, SGDM and RMSProp, to explore the best performing setting. Weight and function space similarities during the optimization trajectory are investigated, and critical differences in the learning process evolution are noted and compared to theory. We observe ADAM optimizer and Glorot initialization consistently performs better, while RMSProp outperforms SGDM as the latter appears to have been stuck at a local minima. Therefore, this autoML BayesOpt framework provides a means to choose the best hyper-parameter settings for a given dataset.

Social media giants like Facebook are struggling to keep up with fake news, in the light of the fact that disinformation diffuses at lightning speed. For example, the COVID-19 (i.e. Coronavirus) pandemic is testing the citizens' ability to distinguish real news from falsifying facts (i.e. disinformation). Cyber-criminals take advantage of the inability to cope with fake news diffusion on social media platforms. Fake news, created as a means to manipulate readers to perform various malicious IT activities such as clicking on fraudulent links associated with the fake news/posts. However, no previous study has investigated the strategies used to create fake news on social media. Therefore, we have analysed five data-sets using Machine Learning (ML) that contain online news articles (i.e. both fake and legitimate news) to investigate strategies of creating fake news on social media platforms. Our study findings revealed a threat model understanding strategies of crafting fake news which may highly likely diffuse on social media platforms.

Segmentation of Magnetic Resonance Images (MRI) of abdominal organs is useful for analysis prior to surgical procedures and for further processing. Deep Learning (DL) has become the standard, researchers have proposed improvements that include multiple views, ensembles and voting. Loss function alternatives, while being crucial to guide automated learning, have not been compared in detail. In this work we analyze limitations of popular metrics and their use as loss, study alternative loss variations based on those and other modifications and search for the best approach. An experimental setup was necessary to assess the alternatives. Results for the top scoring network and top scoring loss show improvements between 2 and 11 percentage points (pp) in Jaccard Index (JI), depending on organ and patient (sequence), for a total of 22 pp over 4 organs, all this being obtained just by choosing the best performing loss function instead of cross-entropy or dice. Our results apply directly to MRI of abdominal organs, with important practical implications for other architectures, as they can be applied easily to any of them. They also show the worth of variants of loss function and loss tuning, with future work needed to generalize and test in other contexts.

Adversarial Classification (AC) is a major subfield within the increasingly important domain of adversarial machine learning (AML). Most approaches to AC so far have followed a classical game-theoretic framework. This requires unrealistic common knowledge conditions untenable in the security settings typical of the AML realm. After reviewing such approaches, we present alternative perspectives on AC based on Adversarial Risk Analysis.

Precision agriculture is a technology used by farmers to help food sustainability amidst growing population. One of the tools of precision agriculture is yield monitoring, which helps a farmer manage his production. Yield monitoring is usually done during harvest, however it could also be done early in the growing season. Early prediction of yield, specifically for fruit trees, aids the farmer in the marketing of their product and assists in managing production logistics such as labor requirement and storage needs. In this study, a machine vision system is developed to estimate fruit yield early in the season. The machine vision system uses a color camera to capture images of fruit trees during the full bloom period. An image segmentation algorithm based on an artificial neural network was developed to recognize and count the blossoms on the tree. The artificial neural network segmentation algorithm uses color information and position as input. The resulting correlation between the blossom count and the actual number of fruits on the tree shows the potential of this method to be used for early prediction of fruit yield.

The novel coronavirus disease (COVID-19) pandemic has impacted health and wellbeing globally. To strengthen preventive and clinical care amid this pandemic, technological innovations like artificial intelligence (AI) are increasingly used in different contexts. This bibliometric study aimed to assess the current scholarly development and prominent research domains in applications of AI technologies in COVID-19 research. A total of 105 articles were retrieved from MEDLINE database that emphasized on the use of AI in the context of COVID-19. Most articles had multiple authors with a collaboration index of 7.18. Moreover, most of the articles were produced from the USA (22.86%) and China (21.9%), whereas developing countries were underrepresented among the contributing nations. Furthermore, several research domains were identified, including prevention and control, diagnostics, epidemiological characteristics, therapeutics, psychological conditions, and different areas of data sciences related to COVID-19. The current bibliometric evidence shows the early stage of development in this field, which necessitates equitable applications of AI in COVID-19 research emphasizing on health disparities, socio-legal issues, vaccine development, and applied public health research in this pandemic.

Long noncoding RNA (lncRNA) are implicated in various genetic diseases and cancer, attributed to their critical role in gene regulation. RNA sequencing is used to capture their transcripts from certain cell types or conditions. For some studies, lncRNA interactions with other biomolecules have also been captured, which can give clues to their mechanisms of action. Complementary \textit{in silico} methods have been proposed to predict non-coding nature of transcripts and to analyze available RNA interaction data. Here we provide a critical review of such methods and identify associated challenges. Broadly, these can be categorized as reference-based and reference-free or \textit{ab initio}, with the former category of methods requiring a comprehensive annotated reference. The \textit{ab initio} methods can make use of machine learning classifiers that are trained on features extracted from sequences, making them suitable to predict novel transcripts, especially in non-model species. Machine learning approaches such as Logistic Regression, Support Vector Machines, Random Forest, and Deep Learning are commonly used. Initial approaches relied on basic sequential features to train the model, whereas the use of secondary structural features appears to be a promising approach for functional annotation. However, adding secondary features will result in model complexities, thus demanding an algorithm that can handle it and furthermore, considerably increasing the utilization of computation resources. Computational strategies combining identification and functional annotation which can be easily customized are currently lacking. These can be of immense value to accelerate research in this class of RNAs.

In this experiment, an efficient and accurate network of detecting automatically disseminated (bot) content on social platforms is devised. Through the utilisation of parallel convolutional neural network (CNN) which processes variable n-grams of text 15, 20, and 25 tokens in length encoded by Byte Pair Encoding (BPE), the complexities of linguistic content on social platforms are effectively captured and analysed. With validation on two sets of previously unexposed data, the model was able to achieve an accuracy of around 96.6% and 97.4% respectively — meeting or exceeding the performance of other comparable supervised ML solutions to this problem. Through testing, it is concluded that this method of text processing and analysis proves to be an effective way of classifying potentially artificially synthesized user data — aiding the security and integrity of social platforms.

In this paper we address the understanding of the problem, of why a deep learning model decides that an individual is eligible for a loan or not. Here we propose a novel approach for inferring, which attributes matter the most, for making a decision in each specific individual case. Specifically we leverage concepts from neural attention to devise a novel feature wise attention mechanism. As we show, using real world datasets, our approach offers unique insights into the importance of various features, by producing a decision explanation for each specific loan case. At the same time, we observe that our novel mechanism, generates decisions which are much closer to the decisions generated by human experts, compared to the existent competitors.

In Schizophrenia, pathway-genotypes may be constructed by combining interrelated immune biomarkers with changes in specific neurocognitive functions that represent aberrations in brain neuronal circuits. These constructs provide insight on the phenome of schizophrenia and show how pathway-phenotypes mediate the effects of genome X environmentome interactions on the symptomatology/phenomenology of schizophrenia. Nevertheless, there is a lack of knowledge how to construct pathway-phenotypes using Partial Least Squares (PLS) path modeling and Soft Independent Modeling of Class Analogy (SIMCA). This paper aims to provide a step-by-step utilization guide for the construction of pathway-phenotypes that reflect aberrations in the neuroimmune - brain circuit axis (NIBCA) in deficit schizophrenia. This NIBCA index is constructed using immune biomarkers (CCL-2, CCL-11, IL-1β, sIL-1RA, TNF-α, sTNFR1, sTNFR2) and neurocognitive tests (Brief Assessment of Cognition in Schizophrenia) predicting overall severity of schizophrenia (OSOS) in 120 deficit SCZ and 54 healthy participants. Using SmartPLS path analysis, a latent vector is extracted from those biomarkers and cognitive tests, which shows a good construct reliability (Cronbach alpha and composite reliability) and replicability and which is reflectively measured through its NIBCA manifestations. This NIBCA pathway-phenotype explains 75.0% of the variance in PHEMN (psychotic, hostility, excitation, mannerism and negative) symptoms. Using SIMCA, we constructed a NIBCA pathway-class that defines deficit schizophrenia as a qualitatively distinct nosological entity and which allows patients with deficit schizophrenia to be authenticated as belonging to the deficit schizophrenia class. In conclusion, our nomothetic approach to develop a nomological network combining neuro-immune and neurocognitive phenome markers to predict OSOS and cross-validate a diagnostic class generated replicable models reflecting the key phenome of the illness, which may mediate the effects of genome X environmentome interactions on the final outcome phenome features, namely symptomatology and phenomenology.

Decision-makers are concerned with the inherent complexity of the modern world's markets. However, price fluctuations, environmental concerns, technological development, emerging markets, political challenges, and social expectations made the 21^st century's more dynamic and complex. From a policy-making perspective, it is vital to uncover future trends. This paper proposed that artificial intelligence can improve interpretations in complex markets, such as financial and energy markets. In a complex environment, it is critical to investigate maximum available input features to ensure no valuable informative feature is neglected. Some AI-based models are investigated and presented that AI-based models can successfully uncover future trends. From a scenario development perspective purified input features subset refer to driving forces which shape alternative futures. Results showed that using AI can improve our understanding of how input features influence future behaviors and simultaneously improves prediction accuracy and reliability.

Sleep postures monitoring systems in the hospital aim at transforming sensing signals into quantitative data to characterize the sleep behaviors of the patient. However, a home-care sleep posture monitoring system needs to be user friendly. In this paper, we present iSleePost - a user-friendly home-care intelligent sleep posture monitoring system. We address the labor-intensive labeling issue of traditional machine learning approaches in the training phase. Our proposed mobile health (mHealth) system leverages the communications and computation capabilities of mobile phones for provisioning a continuous sleep posture monitoring service. Our experiments show that iSleePost can achieve 90 percent accuracy in recognizing sleep postures. More importantly, iSleePost demonstrates that an easily-wear wrist sensor can accurately quantify sleep postures.

In the fashion field, the use of electroplated small metal parts such as studs, clips and buckles is widespread. The plate is often made of precious metal, such as gold or platinum. Due to the high cost of these materials, it is strategically relevant and of primary importance for manufacturers to avoid any waste by depositing only the strictly necessary amount of material. To this aim, Companies need to be aware of the overall number of items to be electroplated so that it is possible to properly set the parameters driving the galvanic process. Accordingly, the present paper describes a Machine Vision-based method able to automatically count small metal parts arranged on a galvanic frame. The devised method relies on the definition of a proper acquisition system and on the development of image processing-based routines. Such a system is then implemented on a counting machine is meant to be adopted in the galvanic industrial practice to properly define a suitable set or working parameters (such as current, voltage and deposition time) for the electroplating machine and, thereby, to assure the desired plate thickness from one side and to avoid material waste on the other.

XGBoost is recognized as an algorithm with exceptional predictive capacity. Models for a binary response indicating the existence of accident claims vs. no claims can be used to identify the determinants of traffic accidents. We compare the relative performances of logistic regression and XGBoost approaches for predicting the existence of accident claims using telematics data. The dataset contains information from an insurance company about individuals’ driving patterns – including total annual distance driven and percentage of total distance driven in urban areas. Our findings show that logistic regression is a suitable model given its interpretability and good predictive capacity. XGBoost requires numerous model-tuning procedures to match the predictive performance of the logistic regression model and greater effort as regards interpretation.

Crop breeding is as ancient as the invention of cultivation. In essence, the objective of crop breeding is to improve plant fitness under human cultivation conditions, making crops more productive while maintaining consistency in life cycle and quality. The applications of predictive breeding has been gaining momentum in agricultural industry and public breeding programs for the last decade, in the aftermath of genomic selection being recognized and widely applied for accelerating genetic gain in breeding programs. The massive amounts of data that has been generated by industry and farmers year after year through several decades has finally been recognized as an asset. A wide range of analytical methods such as machine learning, deep learning and artificial intelligence that were initially developed for diverse quantitative disciplines are now being adopted to crop breeding decision making processes. New technologies are currently being developed that would enable integration of data from various domains such as geospatial variables and a multitude of phenotypic responses as well as genetic information, in order to identify, develop and improve crop faster via partial or full automation of the decisions that pertain to variety development. Here we will discuss and summarize efforts from public and private domains for predictive analytics, and its applications to crop breeding and agricultural product development, and provide suggestions for future research.

In this paper, modification of dual-Halbach permanent magnet (PM) array is investigated to improve performance of tubular linear machine, in terms of flux density and output power. Instead of a classical Halbach array with only radial and axial PMs, proposed model involves insertion of mig-magnets, which have magnetized angle shifted from the reference magnetized angles of axial and radial PMs. This structure leads to elimination of flux leakage and concentration of flux linkage in middle of the coil; therefore, the performance of machine is increased.

Neural microprobes represent an important component of neural prosthetic systems where implanted microprobes record the electro-potentials generated by specific thoughts in the brain and convey the signals to algorithms trained to interpret these thoughts. Here, we present novel elongated multi-site neural probe that can reach depths greater than 10mm. We hypothesize that reaching such depth allows the recording of cognitive signals required to drive cognitive prosthetics. The impedance of the recording sites on the probes was on the order of 500 kΩ at 1 kHz, which is consistent with probes used for neurophysiological recordings. The probes were made porous using Xenon Difluoride (XeF₂) dry etching to improve the biocompatibility and their adherence to the surrounding neural tissue. Numerical studies were performed to determine the reliability of the porous probes. We implanted the elongated probe in rats and show that the elongated probes are capable of simultaneously recording both spikes and local field potentials (LFPs) from various recording sites.

Global food security for the increasing world population not only requires increased sustainable production of food but a significant reduction in pre- and post-harvest waste. The timing of when a fruit is harvested is critical for reducing waste along the supply chain and increasing fruit quality for consumers. The early in field assessment of fruit ripeness and prediction of the harvest date and yield by non-destructive technologies have the potential to revolutionize farming practices and enable the consumer to eat the tastiest and freshest fruit possible. A variety of non-destructive techniques have been applied to estimate the ripeness or maturity but not all of them are applicable for in situ (field or glasshouse) assessment. This review focuses on the non-destructive methods which are promising, or have already been, applied to the pre-harvest in field measurement including colorimetry, visible imaging, spectroscopy and spectroscopic imaging. Machine learning and regression models used in assessing ripeness are also discussed.

This paper presents new above-ground biomass (AGB) and biomass components equations for seventeen forest species in the temperate forests of northwestern Mexico. A data set corresponding to 1336 destructively sampled oak and pine trees was used to fit the models. Generalized method of moments was used to simultaneously fit systems of equations for biomass components and AGB, to ensure additivity. Additionally, the carbon content of each tree component was calculated by the dry combustion method, in a TOC analyser. The fitted equations accounted for on average 91, 83, 84 and 78% of the observed variance in stem wood and stem bark, branch and foliage biomass, respectively, whereas the total AGB equations explained on average 93% of the total observed variance in AGB. The inclusion of h or d²h as additional predictor in the d-only based equations systems slightly improved estimates of stem wood, stem bark and total above-ground biomass, and greatly improved the estimates produced by the branch and foliage biomass equations. The fitted equations were used to estimate AGB stocks at stand level from a database on growing stock from 429 permanent sampling plots. Three machine-learning techniques were used to model the estimated stand level AGB and carbon contents; the selected models were applied to map the AGB and carbon distributions in the study area, which yielded mean values of 129.84 Mg ha^-1 and 63.80 Mg ha^-1, respectively.

Recently, there has been, on the one hand, an increase in the complexity of digital device designs and, on the other hand, an increase in the requirements for the development time and the reliability of the designs. One of the directions of solving this problem is developing new techniques for designing digital devices.This paper proposes a new technique for designing digital devices based on finite state machines with datapath (FSMD), when the functioning of the device is described in the form of an algorithm state machine with datapath (ASMD) charts. The new technique is called ASMD-FSMD. Different digital device design techniques are compared to each other using design examples of a synchronous multiplier on field programmable gate array (FPGA). The efficiency of the ASMD-FSMD technique compared to the traditional approach in terms of area and performance was investigated. The ASMD-FSMD technique, compared to the traditional one, reduces the area from 28.6% to 39.7% and increases the speed for some designs to 17.6%. In addition, using the ASMD-FSMD technique significantly reduces design time and increases design reliability. In conclusion, recommendations for using the ASMD-FSMD technique are made.

The study of hepatitis C virus (HCV) replication in cell culture is mainly based on cloned viral isolates requiring adaptation for efficient replication in Huh7 hepatoma cells. The analysis of wild type (WT) isolates has been enabled by expression of SEC14L2 and by inhibitors targeting deleterious host factors. Here, we aimed at optimizing cell culture models to allow infection with HCV from patient sera. We used Huh7-Lunet cells ectopically expressing SEC14L2, CD81 and a GFP reporter with nuclear translocation upon cleavage by the HCV protease to study HCV replication, combined with a drug-based regimen for stimulation of non-modified wildtype isolates. RT-qPCR based quantification of HCV infections using patient sera suffered from high background in the daclatasvir treated controls. We therefore established an automated image analysis pipeline based on imaging of whole wells and iterative training of a machine-learning tool, using nuclear GFP localization as a readout for HCV infection. Upon visual validation of hits assigned by the automated image analysis the method revealed no background in daclatasvir treated samples. Thereby, infection events were found for 15 of 34 high titer HCV genotype (gt) 1b sera, revealing a significant correlation of serum titer and successful infection. We further show that transfection of viral RNA extracted from sera can be used in this model as well, albeit with so far limited efficiency. Overall, we generated a robust serum infection assay for gt1b isolates using semi-automated image analysis, which was superior to conventional RT-qPCR based quantification of viral genomes.

Climate variations in temperature and precipitation significantly impact forest productivity. Precipitation influences the physiology and growth of species, while temperature regulates photosynthesis, respiration, and transpiration. This study developed bioclimatic models to as-sess how climate change will affect the carbon density of aboveground biomass (cdAGB) in Mex-ico's coniferous forests for 2050 and 2070. We used cdAGB data from the National Forest and Soils Inventory (INFyS) of Mexico and 19 bioclimatic variables from WorldClim ver. 2.0. The best predictors of cdAGB were obtained using machine learning techniques with the 'caret' library in R. The model was trained with 80% of the data and validated with the remaining 20% using Generalized Linear Models (GLM). Current cdAGB prediction maps were generated using the best predictors. Future cdAGB was calculated with the average of three general circulation mod-els (GCMs) of future climate projections from the Coupled Model Intercomparison Project Phase 5 (CMIP5), under four Representative Concentration Pathways (RCPs): 2.6, 4.5, 6.0, and 8.5 W/m². The results indicate cdAGB losses in all climate scenarios, reaching up to 15 Mg C ha-1, and could occur under the RCP 8.5 scenario by 2070 in the central region of the country. Tempera-ture-related variables are more important than precipitation variables. Bioclimatic variables can explain up to 20% of the total variance in cdAGB. The temperature in the study area is expected to increase by 2.66 °C by 2050 and 3.36 °C by 2070, while precipitation is expected to fluctuate by ±10% relative to the current values, which. could geographically redistribute the cdAGB of the country's coniferous forests. These findings underscore the need for forest management to focus not only on biodiversity conservation but also on the carbon storage capacity of these ecosys-tems.

Fusobacterium nucleatum (F. nucleatum, Fn), an oral commensal that can become pathogenic, is associated with periodontitis (PD), adverse pregnancy outcomes, and colorectal cancer (CRC). MicroRNAs (miRNAs) are conserved, non-coding RNAs that play a regulatory role in gene expression and are detected in microbial infections. The aim of this study was to characterize the global microRNA expression kinetics in the mandibles of C57BL/6J mice infected with F. nucleatum for 8 and 16 weeks, and to identify miRNAs as biomarkers of the PD process using high-throughput NanoString nCounter® miRNA panels. Mice were divided into four groups: 8 weeks of Fn infection, 16 weeks of Fn infection, and their respective sham infection. Fn-infected mice at both 8- and 16 weeks showed 100% bacterial colonization on the gingival surface, along with a significant increase in alveolar bone resorption (p<0.0001) and intravascular dissemination to heart, indicating its invasive potential. Out of 577 miRNAs analyzed, seven miRNAs (miR-361-5p, miR-99b-5p) were upregulated, and two miRNAs (miR-362-3p and miR-720) were downregulated in the 8-week Fn infection group. In the 16-week Fn infection group, seven miRNAs (miR-361-5p, miR-99b-5p) were upregulated miRNAs and 13 miRNAs (miR-323-3p, miR-488) were downregulated. Notably, miRNAs such as miR-205, miR-210, miR-199a-3p which were differentially expressed (DE) in the Fn-infected mice at 8 weeks and miR-28 at 16 weeks, have been previously reported in human periodontitis. The 13 miRNAs induced by F. nucleatum (miR-361-5p) are linked to multiple malignancies, including esophageal, gastric, pancreatic, and colorectal cancers. Additionally, the DE miR-126-5p in the 8-week infection group has been identified as a potential biomarker for patients with PD and cardiovascular disease. The results indicate that F. nucleatum acts as a potent oncomicrobe, inducing several miRNAs associated with PD and linking it to systemic comorbidities. Furthermore, the study revealed similar expressions between PD-associated miRNAs and nine CRC tumor-expressed miRNAs.

Recommending products effectively based on historical transaction data and metadata is crucial for improving user satisfaction and increasing sales. This study presents a recommendation system evaluated using Mean Average Precision (MAP@12), leveraging a diverse dataset that includes apparel types, customer demographics, product descriptions, and images. Our approach integrates Mult-Recall methods and LightGBM Ranker, utilizing various recall strategies and advanced feature engineering techniques to enhance recommendation accuracy. By combining these advanced machine learning models and ensemble techniques, our proposed solution outperforms existing methods, providing a robust and efficient recommendation system tailored to diverse customer needs and product characteristics. This integrated approach not only addresses the challenges of data sparsity and cold start but also adapts to the dynamic nature of user preferences and market trends. The results demonstrate significant improvements in both recommendation precision and user satisfaction, highlighting the potential of our approach for real-world e-commerce applications.

In this paper, we conduct a comprehensive analysis of the RounD dataset to enhance the understanding of motion forecasting for autonomous vehicles in complex roundabout environments. We implement a trajectory prediction framework that combines Long Short-Term Memory (LSTM) networks with graph-based modules to model vehicle interactions. Our primary objective is to assess the generalizability of a standard trajectory prediction model across diverse training and testing datasets. Through extensive experiments, we analyze how varying data distributions, including different road configurations and recording times, influence the model’s prediction accuracy and robustness. This study provides key insights into the challenges of domain generalization in autonomous vehicle trajectory prediction.

The Cyber Security Incident Response Team executes its critical duties in the Centralized Security Operation Centers. Its primary target is to protect organizational resources from all types of Advanced Persistent Threats (APT) and attacks while making the correct judgments at the right time. Despite the great efforts and the efficient tools, the incident response mechanism faces two significant challenges. The generation of large amounts of alerts with huge rates of false positives per unit of time and the time-consuming manual expert engagement in the post-alert decision phase. The main aim of this research study is to investigate the potential of Machine Learning techniques in solving the above challenges. This study proposes six event detection modules to identify 14 different types of malicious behavior using Splunk. The available APT dataset is extended by adding more alert types. Then, the machine learning techniques i.e., Random Forest, and XGBoost, are applied to improve the decision process and reduce the false positive alerts of the Security Information and Event Management system. The proposed model outperforms the state-of-the-art techniques by predicting an accuracy of 99.6%.

Automatic Milking Systems (AMS) are revolutionizing the dairy industry by boosting herd effi-ciency, primarily through increased milk yield per cow and reduced labor costs. The performance of milking machines, whether traditional or automated, can be evaluated using advanced vacuum meters through dynamic testing. This process involves scrutinizing the system and milking rou-tine to identify critical points, utilizing the VaDia™ logger (BioControl AS, Rakkestad, NO). Vacuum recordings were downloaded and analyzed using the VaDia Suite™ software under the guidance of a milking specialist. Access to data from AMS across various manufacturers and herds facilitated a retrospective study aimed at describing and comparing key milk emission parameters for different AMS brands while identifying potential mastitis risk factors. Using proper statistical procedures of SPSS 29.1 (IBM Corp, USA), researchers analyzed data from 4,878 individual quarter milkings from cows in 48 dairy herds. Results indicated significant variability in milking pa-rameters associated to quarter milk yield and AMS brand. Notably, despite AMS standardizing teat preparation and stimulation, the study revealed a surprisingly high frequency of two major mastitis risk factors—bimodality and irregular vacuum fluctuations—occurring more frequently than in conventional milking systems. This study, one of the few comparing different AMS brands and their performance, highlights the crucial role of dynamic testing in evaluating AMS perfor-mance under real-world conditions.

This paper examines the utility and exploitation of non-obvious "alien" AI constructions, artificial intelligence solutions that significantly deviate from human intuition and traditional problem-solving approaches. It explores recent revelations, key characteristics, applications across various domains, and far-reaching implications for scientific research and innovation. Particular attention is given to developments in mathematical reasoning, drug discovery, and healthcare. The paper also discusses challenges in implementation, ethical considerations, and future directions, including meta-learning algorithms and human-AI collaboration frameworks. By synthesizing current research and identifying emerging trends, this paper aims to provide a basic understanding of the transformative potential of non-obvious AI solutions in reshaping scientific discovery and biomedical innovation.

Hypertension is a major risk factor for cardiovascular and cerebrovascular diseases. Previous studies have indicated that the presence of heavy metals in the air may be a risk factor for hypertension. This study examined the combined effects of heavy metals on blood pressure considering, sex differences. A health impact survey was conducted with 561 residents near waste incineration facilities in Buki-myeon, Cheongju, Korea. Blood samples were analyzed for heavy metal concentrations, and the blood pressure was measured. Multiple linear and Bayesian kernel machine regressions (BKMR) were used to assess the association between heavy metal exposure and blood pressure. Men had higher blood mercury and lead concentrations, while women had higher cadmium levels. The multiple linear regression analysis revealed no significant associations between heavy metals and systolic blood pressure. However, cadmium was significantly associated with increased diastolic blood pressure in the total group and women, whereas lead was associated with the same in men. In the BKMR analysis, the joint effect of the three metals was significantly associated with diastolic blood pressure below the 45th percentile and above the 55th percentile only in men. These findings underscore the importance of considering sex differences in environmental health studies and public health strategies.

With advancements in remote sensing technology and affordable design, uncrewed aerial systems (UAS), commonly known as drones, have become prevalent in both civil and military applications, such as agriculture, public safety, and aerial imaging. However, the rise in unlawful UAS activities, such as non-compliance with legal standards and potential terrorist attacks, has raised significant public concern, necessitating effective detection and mitigation solutions. Despite the growing importance of this issue, comprehensive and detailed examinations of existing counter-UAS solutions are lacking. To address this gap, this study conducts a bibliometric analysis and scoping review of the current literature to identify key topics and emerging trends in counter-UAS approaches. Utilizing co-word and social network analyses, the study identifies strong and weak connections between selected keywords from academic articles. This study summarizes the limitations and potential opportunities within counter-UAS research, suggesting an increasing focus on multisensory fusion and machine-learning approaches for drone detection and mitigation. Additionally, areas such as swarm drone operations, UAS traffic management (UTM), and UAS networks are identified as important but promising fields for further investigation. The findings of this study provide a foundation for enhancing air and ground safety through improved counter-UAS applications.

In recent years, with the development of the Internet of Things and distributed computing, the "server-edge device" architecture has been widely deployed. This study focuses on leveraging autoencoder technology to address the binary classification problem in network intrusion detection, aiming to develop a lightweight model suitable for edge devices. Traditional intrusion detection models face two main challenges when directly ported to edge devices: inadequate computational resources to support large-scale models and the need to improve the accuracy of simpler models. To tackle these issues, this research utilizes Extreme Learning Machine for its efficient training speed and compact model size to implement autoencoders. Two improvements over the latest related work are proposed: First, to improve data purity and ultimately enhance detection performance, this study partitions the data into multiple regions based on the prediction results of these autoencoders. Second, leveraging autoencoder characteristics to investigate further the data within each region. We used the public dataset NSL-KDD to test the behavior of the proposed mechanism. The experimental results show that when dealing with multi-class attacks, the model's performance was significantly improved, the accuracy and the F1-score are improved by 3.5% and 2.9%, respectively maintaining its lightweight nature.

Explainable Artificial Intelligence (XAI) plays a vital role in increasing transparency and trust in machine learning models, particularly when applied to tabular data which is used in domains such as finance, healthcare, and marketing. This paper presents an extensive survey of XAI techniques used with tabular data and aims to analyze recent research developments since 2021. The survey classes and describes several XAI techniques pertinent to tabular data, it identifies challenges specific to this domain, and explores potential applications and emerging trends. Future research directions are outlined, concentrating on the need for clear definitions of terminology used, data security, user-centric explanations, enhanced interaction, robust evaluation metrics, and advancements in adversarial example-based analysis. The aim is to contribute to the evolving field of XAI, and provide insights for effective, trustworthy, and transparent decision-making using tabular data.

Cellular senescence is the irreversible growth arrest subsequent to oncogenic mutations, DNA damage or metabolic insult. Senescence is associated with aging and chronic age associated diseases such as cardiovascular disease and diabetes. The involvement of cellular senescence in Acute Kidney Injury (AKI) and Chronic Kidney Disease (CKD) is not fully understood however, recent studies suggest that such patients have a higher-than-normal level of cellular senescence and accelerated aging. This study aimed to discover key biomarkers of senescence in AKI and CKD patients compared to other chronic ageing diseases as controls. We show that senescence proteins CKAP4 and PTX3 are upregulated in AKI and CKD patients compared with controls with chronic diseases suggesting they may play in a role in overall kidney disease development. Additionally, CKAP4 was found to be differentially expressed in both AKI and CKD when compared to unhealthy controls hence this biomarker could be prognostic senescence biomarker of both AKI and CKD.

In manufacturing chatter is an unwanted phenomenon that can lead to product quality reduction and tool wear. Real time chatter detection is key to preventing these issues and improving overall machining efficiency. In this paper we propose an unsupervised chatter detection method using autoencoders (AE) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm that uses internal signals of Computer Numerical Control (CNC) machines. The proposed method starts by using an AE to extract features from raw internal signals collected from CNC machines. This step reduces the dimensionality of the data and captures the underlying patterns of chatter. Then the extracted features are fed into DBSCAN clustering algorithm which is a density based algorithm that groups similar data points and identifies outliers. We tested the proposed method with real world data collected from various CNC machines. The results show that our unsupervised chatter detection method has high accuracy, precision and recall, can detect chatter and distinguish it from normal machining. Also the method is robust to noise and can adapt to dynamic machining conditions. In summary our work presents an unsupervised chatter detection method using AE and DBSCAN clustering that uses internal signals of CNC machines. This method is a reliable and efficient solution for real time chatter detection so manufacturers can improve product quality, optimize machining process and reduce tool wear during machining.

The exponential growth of multimodal content across social media platforms, comprising text, images, audio, and video, has catalyzed substantial interest in artificial intelligence, particularly in multi-modal sentiment analysis (MSA). This study presents a comprehensive survey of 30 research papers published between 2020 and 2024 by eminent publishers such as Elsevier, ACM, IEEE, Springer, and others indexed in Google Scholar. Our analysis primarily focuses on exploring multimodal fusion techniques and features, with specific emphasis on the integration of text and image data. Additionally, the article offers an overview of the evolution, definition, and historical context of MSA. It delves into the current challenges and potential advantages of MSA, investigating recent datasets and sophisticated models. Furthermore, the study provides insights into prospective research directions. Notably, this review offers valuable recommendations for advancing research and developing more robust MSA models, thus serving as a valuable resource for both academic and industry researchers engaged in this burgeoning field.

Procedural and reporting guidelines frame the process of scientific practices and communications among researchers and the community at large. In the pursuit of fostering reproducibility, several methodological frameworks have been proposed by several initiatives. Nevertheless, recent studies indicate that data leakage and reproducibility are still prominent challenges. Recent studies have shown the transforming potential of incorporating the FAIR (Findable, Accessible, Interoperable and Reusable) principles in the work-flow of different context such as software and machine learning model development stages to cultivate open science. In this work, we introduce a framework to calibrate reporting guidelines against the FAIR principles in order to foster reproducibility and open science. We adapted the “Best fit” framework synthesis approach to develop the calibration framework. We propose a series of defined workflows to calibrate reporting guidelines with FAIR principles and a use case to demonstrate the process. By integrating FAIR principles with established reporting guidelines, the proposed framework bridges the gap in accommodating both FAIR metrics and reporting guidelines and benefits from advantages of these major integrated components.

The proliferation of Internet of Things (IoT) devices on Linux platforms has heightened concerns regarding vulnerability to malware attacks. This paper introduces a novel approach to investigating the behavior of Linux IoT malware by examining syscalls and library syscall wrappers extracted through static analysis of binaries, as opposed to the conventional method of using dynamic analysis for syscall extraction. We rank and categorize Linux system calls based on their security significance, focusing on understanding malware intent without execution.Feature analysis of the assigned syscall categories and risk ranking is conducted with statistical tests to validate their effectiveness and reliability in differentiating between malware and benign binaries. Our findings demonstrate that potential threats can be reliably identified with an F1 score of 96.86%, solely by analyzing syscalls and library syscall wrappers. This method can augment traditional static analysis, providing an effective preemptive measure to enhance Linux malware analysis. This research highlights the importance of static analysis in strengthening IoT systems against emerging malware threats.

In Kenya, reports of Rift Valley fever (RVF), one of the worst climate-sensitive zoonosis, have been common. Despite the fact that several empirical studies have demonstrated that Machine learning techniques perform better than time series models in forecasting time series data, there is little evidence of their use in predicting disease outbreaks in Africa. Recently, the literature has mentioned a number of other uses of machine learning to support intelligent decision-making in the healthcare industry and public health but there is limited knowledge on the use of the XGBoost model in the prediction of disease outbreaks. Among the Kenyan provinces, Rift valley fever cases were more pronounced in Rift valley (26.80%) and Eastern (20.60%) regions. The study explored the relationship be-tween RVF incidence and various climatic factors, such as humidity, clay content, elevation, slope, and rainfall. The strongest correlation, a meager 0.02903 for rainfall, was found in the correlation matrix, which showed weak linear relationships between various climatic factors and RVF cases. These climate variables were used to train the XGBoost model, which showed remark-able performance with an AUC of 0.8908, accuracy of 99.74%, precision of 99.75%, and recall of 99.99%. Rainfall was found to be the most important predictor in the feature importance analysis. These findings are consistent with other research showing how important weather conditions are to RVF outbreaks. According to the study's findings, the use of sophisticated machine learning models that take a variety of climatic factors into account can greatly improve RVF outbreak prediction and control.

NMR spectroscopy, which is based on the phenomenon of nuclear magnetic resonance, has been widely popularized by its application in medical imaging under the name of MRI. In the organic laboratory, the 13C NMR spectrum of a newly synthetized compound remains an essential step in elucidating its structure. For the chemist, the interpretation of such a spectrum, which is a set of chemical shift values, is made easier if he has a tool capable of predicting with sufficient accuracy the carbon shift values from the structure he intends to prepare. As there are few open source methods for accurately estimating this property, we applied our graph machine approach to build models capable of predicting the chemical shifts of carbons. For this study, we have focused on benzene compounds building an optimized model derived from training a database of 10577 chemical shifts originating from 2026 structures which contain up to ten types of atoms other than carbon, namely H, O, N, S, P, Si and halogens. It provides a training root mean squared relative error (RMSRE) of 0.5 %, i.e. a root mean squared error (RMSE) of 0.6 ppm, and a mean absolute error (MAE) of 0.4 ppm for estimating the chemical shifts of the 10k carbons. The predictive capability of the graph machine model is also compared with that of three commercial software on a data set of 171 original benzenic structures (1012 chemical shifts). The graph machine model proves very efficient in predicting chemical shifts with an RMSE of 0.9 ppm, and compares favorably with the RMSEs of 3.4, 1.8 and 1.9 ppm computed with ChemDraw, ACD and MestReNova softwares respectively. Finally, a Docker-based tool is proposed to predict the carbon chemical shifts of benzenic compounds solely from their SMILES codes.

This study introduces an integrated method for managing process risks in a Business Process Reengineering (BPR) project, using Robust Data Envelopment Analysis (RDEA) and machine learning (ML). The goal is to prioritize risks based on three standard factors of PFMEA: Severity, Occurrence, and Detection (S-O-D), and incorporating two additional factors (Breakdown Cost and Breakdown Duration) seen as undesirable outputs. The model also accounts for the effect of uncertainty on expert-estimated values by applying disturbance percentages in the linear PFMEA-RDEA model. A machine learning model is proposed to predict new values if partial or total modifications have been made to the processes. The approach was implemented in an au-tomotive sector company, and the results showed the impact of uncertainty on values by com-paring different approaches such as RPN, PFMEA-DEA, and PFMEA-RDEA. A new reduced risk categorization was achieved, who allowed for decision-makers to focus on necessary actions for reengineering.

This work focuses on short-term forecasting of PV output power. The multilayer perception (MLP), convolutional neural networks (CNN), and k-nearest neighbour (kNN) neural networks have been used singly or in a hybrid (with other algorithms) to forecast solar PV power or global solar irradiance with good success. Their performances in forecasting PV power have been compared with other algorithms but not with themselves. The study aims to compare performance of a number of neural network algorithms in solar PV energy yield forecasting under different weather conditions. The performance of MLPNN, CNN and kNN are being compared using solar PV (hourly) energy yield data for Grahamstown, Eastern Cape, South Africa. The choice of location is part of the study parameters to provide insight into renewable energy power integration in specific areas in south Africa that may be prone to extreme weather conditions. The kNN algorithm was found to have RMSE value of 4.95% and MAE value of 2.74% at its worst performance, and RMSE value of 1.49% and MAE value of 0.85% at its best performance. It outperformed the others by a good margin and kNN could serve as a fast, easy, and accurate tool to forecast solar PV output power.

Although composting has many advantages in the treatment of organic waste, there are still many problems and challenges associated with emissions, like NH3, VOCs, and H2S, as well as greenhouse gases such as CO2, CH4, and N2O. One promising approach to enhancing composting conditions is used of novel analytical methods bad on artificial intelligence. To predict and optimize the emissions (CO, CO2, H2S, NH3) during composting process kinetics thought mathematical models (MM) and machine learning (ML) models were utilized. Data about everyday emissions from laboratory composting with compost’s biochar with different incubation (50, 60, 70 °C) and biochar doses (0, 3, 6, 9, 12, 15% d.m.) were used for MM and ML models selections and training. MM has not been very effective in predicting emissions, (R2 0.1 - 0.9), while ML models such as acritical neural network (ANN, Bayesian Regularized Neural Network; R2 accuracy CO:0,71, CO2:0,81, NH3:0,95, H2S:0,72)) and decision tree (DT, RPART; R2 accuracy CO:0,693, CO2:0,80, NH3:0,93, H2S:0,65) have demonstrated satisfactory results. For the first time the ML models to predict CO and H2S during composting were demonstrated. Further research in a semi-scale and field study composting with biochar is needed to improve the accuracy of developments models.

The dawn of Artificial Intelligence (AI) in healthcare stands as a milestone in medical innovation. A plethora of different medical fields are heavily involved and pediatric emergency medicine is no exception. These new tools do not merely provide more advanced and efficient systems for patients’ diagnosis, management and treatment. They rather concern a strict shift from traditional methods based upon broad categories towards a more personalized healthcare. AI offers many promises in pediatric emergency medicine with a wide range of applications involving clinical decision making, patients’ flows management and prioritization. Main barriers to a widespread diffusion involve technological challenges but also ethical issues and the paucity of extensive datasets in pediatric contexts. We conducted a narrative review structured in two parts. The first part explores the theoretical principles of AI, providing all the necessary background to feel confident with these new state-of-the-art tools. The second part presents an informative analysis of AI models in pediatric emergencies, pointing out the actual applications and challenges until future feasible research perspectives.

The opportunities and advantages of implementing Artificial Intelligence (AI) and Machine Learning (ML) to optimize the process control of wastewater treatment in the Oil and Gas industry are discussed in this work. Discussion is centered around five major points: 1) Various types of Oil and Gas wastewater (Sour, Oily, Produced) and industry standards/methods for treatment, including specific case studies and challenges when treating certain wastewater. This section aims at the common types of wastewater in the Oil and Gas industry and how they are generally treated. 2) How AI/ML has been implemented in wastewater treatment in various industries; summary of which AI/ML languages were used along with a brief overview of how they work. This article focused on an understanding of the specific case studies and AI explanations to understand the effectiveness of wastewater treatment using AI/ML. 3) AI that is currently implemented in Oil and Gas such as personal protective equipment (PPE), smartwatches radiofrequency identification etc. This showed that AI/ML already plays a successful role in the Oil and Gas industry and has proven to help optimize many processes and components. Many of these AI/ML implementations are now industry standards and have greatly benefitted the industry. This section demonstrates that the implementation and the use of AI/ML have been proven to be a successful endeavor and that expanding the role of AI/ML would be beneficial. 5) Using the established points from the previous sections the paper outlines why AI/ML is ideal for Oil and Gas wastewater treatment; this argument centers around fit-for-purpose treatment, and the possible uses of the treated water. A determination of the best-suited AI languages, possible implementation scenarios, and how this would shift the way Oil and Gas industries treat and approach wastewater is presented to the reader.

Deep learning is widely used in image segmentation, effectively identifying crops and weeds and contributing to the reduction of herbicide use. Traditional methods of applying pesticides over large areas are inefficient in dealing with weeds as needed, leading to massive waste of pesticides, high-cost production, and serious environmental pollution. This affects crop yield and quality. Although weed recognition using conventional deep learning methods has evolved over time, there are still challenges in weed extraction, detection, and segmentation. Accurate recognition and detection of weeds are essential prerequisites for implementing variable spraying. This paper proposes a semantic segmentation method based on UNet++ for complex environments where accurate identification of plants and weeds is challenging, targeting weeds in sugar beets, peas, and rice. An attention module is integrated into the upsampling process of UNet++, and UNet++ is used as a backbone network to effectively integrate multi-scale information, efficiently suppressing external noise interference. The UNet++ model with an integrated attention mechanism module achieves higher IOU than the general UNet++ model used in medical image analysis. This method effectively detects crops and weeds in complex backgrounds, providing reference material for the accurate application of robotic herbicides.

The prediction of hydrological flows (rainfall-depth or rainfall-discharge) is becoming increasingly important in the management of hydrological risks such as floods. In this study, the Long Short-Term Memory (LSTM) network, a state-of-the-art algorithm dedicated to time series, is applied to predict the daily water level of lake Nokoue in Benin. This paper aims to provide an effective and reliable method enable of reproducing the future daily water level of Lake Nokoue, which is influenced by a combination of two phenomena: rainfall and river flow (runoff from the Ouémé River, the Sô River, the Porto-Novo lagoon, and the Atlantic Ocean). Performance analysis based on the forecasting horizon indicates that LSTM can predict the water level of Nokoué Lake up to a forecast horizon of t+10 days. Performance metrics such as Root Mean Square Error (RMSE), coefficient of correlation (R²), Nash-Sutcliffe Efficiency (NSE), and Mean Absolute Error (MAE) agree on a forecast horizon of up to t+3 days. The values of these metrics remain stable for forecast horizons of t+1 days, t+2 days, and t+3 days. The values of R² and NSE are greater than 0.97 during the training and testing phases in the Nokoué Lake basin. Based on the evaluation indices used to assess the model's performance for the appropriate forecast horizon of water level in the Nokoué lake basin, the forecast horizon of t+3 days is chosen for predicting future daily water levels.

INTRODUCTION : In this scoping review, we delve into the transformative potential of artificial intelligence (AI) in addressing challenges inherent in whole genome sequencing (WGS) analysis, with a specific focus on its implications in surgical oncology. METHODS: Scoping review of whole genomic sequencing and artificial intelligence.DISCUSSION : Unveiling the limitations of existing sequencing technologies, the review illuminates how AI-powered methods emerge as innovative solutions to surmount these obstacles. The evolution of DNA sequencing technologies, progressing from Sanger sequencing to next-generation sequencing, sets the backdrop for AI's emergence as a potent ally in processing and analyzing the voluminous genomic data generated by these technologies. Particularly, deep learning methods play a pivotal role in extracting knowledge and discerning patterns from the vast landscape of genomic information. In the context of oncology, AI-powered methods exhibit considerable potential across diverse facets of WGS analysis, including variant calling, structural variation identification, and pharmacogenomic analysis. CONCLUSIONS : This review underscores the significance of multimodal approaches in diagnoses and therapies, highlighting the imperative for ongoing research and development in AI-powered WGS techniques. Integrating AI into the analytical framework empowers scientists and clinicians to unravel the intricate interplay of genomics within the realm of multi-omics research, paving the way for more personalized and targeted treatments in surgical oncology and perhaps beyond.

To ensure consumers can purchase high-quality rice, accurate identification of rice varieties is particularly important. Methods: This article conducts research based on machine learning algorithms from the perspectives of image, spectrum, and spectrogram fusion. Six types of hyperspectral image data of rice were preprocessed using convolutional smoothing (SG) and multiple scatter correction (MSC). Texture information of the images was extracted using gray-level co-occurrence matrix. Spectra, texture, and spectrogram data were fused into a new matrix. Spectral, texture, and spectrogram fusion data were used as inputs for the model, and support vector machines (SVM), logistic regression (LR), and K-nearest neighbors (KNN) classification models were constructed and compared. Results: From the classification results, the spectrogram fusion classification performance was better than classification models using only spectra or texture. Conclusion: The research results showed that the accuracy of SVM and LR classification models exceeded 90%, and the LR model performed the best, effectively classifying rice varieties.

Heavy metal pollution poses a significant threat to ecosystems and human health worldwide. In response to this pressing issue, bioremediation has emerged as a promising approach for mitigating contamination. This paper adopts a bibliographic method to explore the key techniques and applications of bioremediation. Furthermore, we delve into the most prominent threats to human environmental health and the corresponding remediation methods. Additionally, we discuss the future trajectory of bioremediation research, with a particular focus on the integration of big data and machine learning technologies. These advanced methodologies hold great potential for enhancing the effectiveness and efficiency of environmental remediation efforts in the face of escalating pollution challenges.

A compositional data vector is a special type of multivariate observation in which the elements of the vector are non-negative and sum to a constant, usually taken to be unity. A compositional data does not have zero and only retains relative information. Furthermore, comparisons can only be made between compositional data of the same component. The relevant sample space is the standard simplex. A simplex space is a space that is a generalized form of a triangle. For compositional data, many of the operations defined in Euclidean space are meaningless. Microbiome analyzes have become popular in recent years. Operational Taxonomic Units (OTU) and Amplicon Sequence Variants (ASVs) used in microbiome analysis are one type of compositional data. The microbiome data are counts of different species within a sample and it is compositional. Although there is an order of bacterial abundance within a sample, there is no order of bacterial abundance between samples. Firmicutes and Bacteroidetes, the major phyla in the colon, have been observed in humans worldwide. Gut microbiome analyses often use the Firmicutes/Bacteroidetes ratio and principal coordinates analysis (PCoA). Alpha and beta diversities are used as indicators of bacterial flora diversity. However, misinformation is pervasive in the human microbiome literature and analysis. There is a lot of misunderstanding regarding compositional data and its analysis. An attempt was made to demonstrate how to analyze using the “National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN JMD) (Public Data).” The results showed that PCoA did not work, and principal component analysis (PCA) was useful for analyzing the gut microbiome relative diversity.

The paper concerns with the identification of current communication technologies and software functionalities from the field of digitalization and Industry 4.0, from the point of view of benefits and applicability for the machine manufacturer. In order to identify the relevant technologies and functionalities, both a descriptive and statistical literature research was carried out. As a result, both the currently most widespread technologies and eleven IT functionalities relevant to mechanical engineers were identified and described in terms of their application and implementation. In addition, a knowledge gap in the area of industry transfer in the field of Industry 4.0 / Digitalization was identified

Machine learning (ML) based on neural network (NN) facilitates data-driven techniques for handling large amounts of data, either obtained through experiments or simulations at multiple spatio-temporal scales, thereby finding the hidden patterns underlying these data and promoting efficient research methods. The main purpose of this paper is to extend the capabilities of a new solver, called realFluidReactingNNFoam, under development at University of Perugia in OpenFOAM with NN algorithm for replacing complex real-fluid thermophysical property evaluations, using the approach of coupling OpenFOAM and Python-trained NN models. Currently, NN models are trained against data generated using the Peng-Robinson equation of state (PR-EoS) assuming mixture frozen temperature. The OpenFOAM solver, where needed, calls the NN models in each grid-cell with appropriate inputs, and the returned results are used and stored in suitable OpenFOAM data structures. Such inference for the thermophysical properties is achieved via the “Neural Network Inference in C made Easy” (NNICE) library, which proved to be very efficient and robust. The overall model is validated considering a liquid-rocket benchmark comprised of liquid-oxygen (LOX) and gaseous-hydrogen (GH2) streams. The model accounts for real-fluid thermodynamics and transport properties, making use of the PR-EoS and the Chung transport models. First, the development of a real-fluid model (RFM) with artificial neural network is described in detail. Then, the numerical results of the transcritical mixing layer (LOX/GH2) benchmark are presented and analyzed in terms of accuracy and computational ef-ficiency. Results of the overall implementation indicate that the combined OpenFOAM and ML approach provides a speed-up factor higher than seven, while preserving the original solver accuracy.

Convolutional neural networks (CNNs) have proven to be a very efficient class of machine learning (ML) architectures for handling multidimensional data by maintaining data locality, especially in the field of computer vision. Data pooling, a major component of CNNs, plays a crucial role for extracting important features of the input data and downsampling its dimensionality. Multidimensional pooling, however, is not efficiently implemented in existing ML algorithms. In particular, quantum machine learning (QML) algorithms have a tendency to ignore data locality for higher dimensions by representing/flattening multidimensional data as simple one-dimensional data. In this work, we propose using the quantum Haar transform (QHT) and quantum partial measurement for performing generalized pooling operations on multidimensional data. We present the corresponding decoherence-optimized quantum circuits for the proposed techniques along with their theoretical circuit depth analysis. Our experimental work was conducted using multidimensional data, ranging from 1-D audio data, to 2-D image data, to 3-D hyperspectral data, to demonstrate the scalability of the proposed methods. In our experiments, we utilized both noisy and noise-free quantum simulations on a state-of-the-art quantum simulator from IBM Quantum. We also show the efficiency of our proposed techniques for multidimensional data by reporting the fidelity of results.

The effect of climate change plays a significant role on groundwater level variations and crop yield. The change in climate leading to increased temperatures, decreased rainfalls and extreme drought conditions ultimately cause low groundwater level and crop yield. Judicious groundwater management technology viz., water recharge options, need based irrigation, crop selection, must be adapted for further increasing the water table level in soil ecosystem. Additionally, climate change impacts combined with changes in agricultural water use can affect groundwater dynamics. Increased irrigation demand and decreased summer precipitation can lower groundwater levels, impacting crop production. Overall, climate change affects both groundwater resources and crop yield highlighting the need for sustainable water management practices and consideration of soil properties in agricultural modelling. The authors carefully selected relevant research articles addressing the impact of climate change on ground water level in Asian countries. The review highlights the use of machine learning method to ground water level and crop yield modelling. According to the study, machine learning techniques have made significant contributions to predicting groundwater level changes and crop yield with higher accuracies, high performance and less running time.

The World Health Organization (WHO) has released reports indicating that heart disorders hold the unfortunate distinction of being the primary cause of death worldwide. Shockingly, an astonishing estimated 17.9 million lives are claimed by heart diseases annually, accounting for an alarming 31% of all global deaths. With all the flaws in the clinical situation, it is frequently challenging to assess the severity of cardiac disease and forecast its course of progression due to heterogeneity and complex interplay of factors. Therefore, early heart disease detection is essential for effective therapy. To address these challenges, Machine Learning (ML) boosting algorithms play a pivotal role as the main components of predictive analytics required to do this. The main objective of this study is to develop a comprehensive comparative framework to predict heart diseases using state-of-the-art machine learning with boosting techniques such as Decision Tree, Random Forest, Gradient Boosting, Catboost, XGboost, Light GBM, and Adaboost. To evaluate the performance of the models, a large heart disorders dataset is used from the UCI machine learning repository, comprised of 26 feature-based numerical and categorical attributes with 8763 samples from all over the globe. The Experimental results reveal that AdaBoost attained the highest accuracy of 95% and outperforms other algorithms concerning various performance measures like precision= 0.98, recall= 0.95, specificity= 0.95, f1-score=0.01, Negative predicted value= 0.83, False positive rate= 0.04, False negative rate= 0.04 and False Development rate= 0.01.

Through the refinement of approaches to applying visual augmentation to virtual environments, the possibilities of utilising this process in robotics continues to be explored. However these endeavours have only yielded a collaborative relationship between the two technologies, where the augmented reality (AR) component is separate from the robot functionality. Regarding telepresence robots, research has shown that their perceived usefulness wanes in comparison to that of video conferencing tools as they do not offer a level of immersion and modified visual experience to justify their use. Therefore this paper contributes a novel virtually augmented robot (VAR) model that embeds AR functionality into telepresence, through the use of augmentation and facial recognition sub-models and algorithms. Experimental results show that using pulse width modulation (PWM), duty cycles, digital signals and accelerometer angles provides an approach for stabilising robot motion and the transference of human to robot head movement. The results also show the benefit of having a robot model with intrinsic AR functionality, as it offers a greater level of immersion and visual control of an environment.

This study utilizes the Random Forest technique to investigate risk management practices and concerns in Islamic banks using survey data from 2016 to 2021. Findings reveal that larger banks provide more consistent survey responses, driven by their confidence and larger survey budgets. Moreover, a positive link is established between a country's development, characterized by high GDPs and low inflation and interest rates, and the precision of Islamic banks' survey responses. Analyzing risk-related concerns, the study notes a significant reduction in credit portfolio risk, attributed to improved risk management practices, global economic growth, stricter regulations, and diversified asset portfolios. Concerns related to terrorism financing and cybersecurity risks have also decreased due to better enforcement of anti-money laundering regulations and investments in cybersecurity infrastructure and education. This research enhances our understanding of risk management in Islamic banks, highlighting the impact of bank size and country development. Additionally, it emphasizes the need for ongoing analysis beyond 2021 to account for potential COVID-19 effects and evolving risk management and regulatory practices in Islamic banking.

Human Action Recognition is widely used in multiple fields to recognize activities and extract spatial and temporal information. This paper focuses on analyzing various methods as well as provides extensive knowledge of the foundational concepts of the HAR. A dataset is also very crucial for any research study so we discussed the popular dataset and its features in this paper.

The prevalence of chronic diseases is dramatically increasing demand for emergency healthcare. Existing systems rely on patients self-identifying symptoms, causing dangerous delays. This study develops an AI and IoT-powered “digital twin” solution to enable continuous real-time monitoring and timely prediction of diverse medical emergencies. A digital twin is a virtual representation of an individual, modeled using multidimensional physiological data from wearable sensors. Machine learning techniques analyze patterns in this data to identify anomalies and predict emergencies like heart attacks or falls. A key contribution is an optimized ensemble algorithm combining gradient boosted trees, neural networks, and other techniques to accurately detect emergency events. Evaluation on a dataset of 9158 samples shows the digital twin identifies key emergencies with over 90% recall, enabling prevention and rapid response. It allows risk stratification and personalized interventions based on early warnings, circumventing over 2 million avoidable emergency room visits annually. This study demonstrates the feasibility of an integrated, predictive, patient-centric emergency response system enabled by digital twin technology.

This study presents a cost-effective IoT-based remote monitoring system for solar PV energy systems, along with a machine learning-based PV power estimator. Remote access is crucial for tracking PV systems installed in remote areas. In this system, an open-source and IoT-compatible data logger is employed. The data logger collects important performance data of the PV system and transfers it to a server. Real-time visualization of this data is displayed in the designed web and mobile monitoring interfaces. The measured data includes the current, voltage, and temperature information of the PV generator and battery, as well as environmental parameters such as temperature, radiation, humidity, and pressure. Subsequently, this data is used for PV power estimation using machine learning methods. This enables the identification of maintenance requirements and the prediction of potential issues in the PV system. When a problem occurs in the PV system, the user is alerted through the mobile application. Early detection and intervention prevent power loss and damage to system components. When comparing the results of linear regression, SVM, decision trees, random forests, and KNN machine learning methods for power estimation based on performance evaluation criteria, it was observed that the random forests algorithm provided the best results. In conclusion, the developed monitoring and estimation system, along with web and mobile interfaces, is suitable for large-scale PV energy systems.

The definition of artificial intelligence (AI) is the capacity of a computer or machine to mimic or simulate human intelligence. AI has been applied in different areas, including medicine. In cardiology, AI techniques have revolutionized the field of cardiac diagnosis, enabling the detection and prediction of cardiovascular diseases to be more accurate and efficient. This article provides an overview of machine learning, deep learning, and neural networks as they pertain to cardiac diagnosis. We will investigate how these techniques harness the power of data and algorithms to analyze complex patterns in cardiac data, thereby facilitating early detection, risk assessment, and treatment decision-making.

Generative artificial intelligence (AI), in particular large language models such as ChatGPT have reached public consciousness with a wide-ranging discussion of their capabilities and suitability for various professions. The extant literature on the ethics of generative AI revolves around its usage and application, rather than the ethical framework of the responses provided. In the education sector, concerns have been raised with regard to the ability of these language models to aid in student assignment writing with the potentially concomitant student misconduct of such work is submitted for assessment. Based on a series of ‘conversations’ with multiple replicates, using a range of discussion prompts, this paper examines the capability of ChatGPT to provide advice on how to cheat in assessments. Since its public release in November 2022, numerous authors have developed ‘jailbreaking’ techniques to trick ChatGPT into answering questions in ways other than the default mode. While the default mode activates a safety awareness mechanism that prevents ChatGPT from providing unethical advice, other modes partially or fully bypass the this mechanism and elicit answers that are outside expected ethical boundaries. ChatGPT provided a wide range of suggestions on how to best cheat in university assignments, with some solutions common to most replicates (‘plausible deniability,’ language adjustment of contract written text’). Some of ChatGPT’s solutions to avoid cheating being detected were cunning, if not slightly devious. The implications of these findings are discussed.

Machine learning methods have been a standard approach to select features that are associated with an outcome and build a prediction model when the number of candidate features is large. LASSO has been one of the most popular approaches to this end. LASSO approach selects features with large regression estimates, rather than based on statistical significance, associating the outcome, by imposing L1-norm penalty to overcome the high dimensionality of the candidate features. As a result, LASSO may select insignificant features while possibly missing significant ones. Furthermore, from our experience, LASSO has been found to select too many features. By selecting features that are not associated with the outcome, we may have to spend more cost to collect and manage them in the future use of a fitted prediction model. Using the combination of L1- and L2-norm penalties, elastic net (EN) tends to select more features than LASSO. The overly selected features that are not associated with the outcome act like white noise, so that the fitted prediction model loses the prediction accuracy. In this paper, we propose to use the standard regression methods (without any penalizing approach) with stepwise variable selection procedure to overcome these issues. Unlike LASSO and EN, this method selects features based on statistical significance. Through extensive simulations, we show that this maximum likelihood estimation based method selects very small number of features while maintaining a high prediction power, while LASSO and EN make a large number of false selections to result in loss of prediction accuracy. Contrary to LASSO and EN, the regression methods combined with a stepwise variable selection method is a standard statistical method, so that any biostatistician can use it to analyze high dimensional data even without advanced bioinformatics knowledge.

The importance of performance excellence and operator’s safety is fundamental not only when operators perform repetitive and controlled industrial tasks, but also in case of abrupt gestures due to inattention and unexpected circumstances. Since optical systems work at too low frequencies and they are not able to detect gestures as early as possible, combining the use of wearable magneto-inertial measurement units (MIMUs) with the adoption of deep learning techniques can be useful to instruct the machine about human motion. To improve the initial training phase of neural networks for high classification performance, gestures repeatability over time has to be verified. Since test-retest approach has been poorly applied based on MIMUs signals in a context of human-machine interaction, the aim of this work was to evaluate the repeatability of pick-and-place gestures composed of both normal and abrupt movements. Overall, results demonstrated an excellent test-retest repeatability for normal movements and a fair-to-good test-retest repeatability for abrupt movements. Moreover whereas the reduction of time between test and retest sessions increased repeatability indices.. In detail, results of the test suggested how to improve the reinforcement learning for the identification of gestures onset, whereas results of the retest gave important information about the time necessary to retrain the network.

Diabetes becomes a life threatening non-communicable disease in the world, according to International Diabetic Federation (IDF) in 2023, an estimated 537 million adults (20-79 years) are living with diabetes, which is equivalent to 9.3% of the global adult population. This number is predicted to rise to 643 million by 2030 and 783 million by 2045. Over 3 in 4 adults with diabetes live in low- and middle-income countries. Diabetes is a persistent metabolic condition marked by increased levels of glucose in the bloodstream. It is a significant global health concern, affecting millions of individuals worldwide. It having the symptoms of drowsiness for the whole day if not properly examined or treated well. It is targeted to spread over the younger age community and the number is growing in the exponential fashion. Even day to day many advancements have come from the researcher to diagnose, to find solutions to prevent from newer entry. Authors have taken a dataset with 57 non diabetic and 20 diabetic patients with the total 28735 micro array gene to undergone pre-processing process and reduced up to 22960 gene data using Dimensionality Reduction (DR) such as Detrend Fluctuation Analysis (DFA), Chi square probability density function (Chi2PDF), Firefly algorithm Cuckoo search were used in this research. Meta heuristic algorithms like Particle swarm Optimization (PSO) and Harmonic Search (HS) are used for feature selection. Further seven classification techniques such as Non-Linear Regression (NLR), Linear Regression (LR), Logistics Regression (LoR), Gaussian Mixture Model (GMM), Bayesian Linear Discriminant Classifier (BLDC), Softmax Discriminant Classifier (SDC), Support Vector Machine – Radial Basis Function (SVM-RBF) are using to make a decision, predictive analysis and segregate the data according to the level of blood glucose as Diabetic Patient (DP) and Non-Diabetic Patient (NDP).

Background: Functional Movement Screening (FMS) allows for rapid assessment of an individual’s physical activity level and timely detection of sports injury risk. However, traditional functional movement screening often requires on-site assessment by experts, which is time-consuming and prone to subjective bias. Therefore, the study of automated functional movement screening has become increasingly important. Methods: In this study, we propose an automated assessment method for FMS based on the improved Gaussian Mixture Model (GMM). First, the oversampling of minority samples is conducted, the movement features are manually extracted from the FMS dataset collected with two Azure Kinect depth sensors, then we train the Gaussian mixture model with different scores (1 point, 2 points, 3 points) of feature data separately, finally, we conducted FMS assessment by the Maximum Likelihood estimation. Results: The improved GMM has a higher scoring accuracy (Improved GMM:0.8) compared to other models (Traditional GMM=0.38, Adaboost.M1=0.7, Naïve-Bayes=0.75), and the scoring results of improved GMM have a high level of agreement with the expert scoring (kappa=0.67). Conclusions: The results show that the proposed method based on the improved Gaussian mixture model can effectively perform the FMS assessment task and it is potentially feasible to use depth cameras for FMS assessment.

Multimodal user interfaces promise a natural and intuitive human machine interactions. But is the extra effort for the development of a complex multi-sensor system justified, or can users also be satisfied with one input modality already? This study investigates interactions in an industrial weld inspection workstation. Three unimodal interfaces, including spatial interaction with buttons augmented on a workpiece or a worktable, and speech commands, were tested individually and in a multimodal combination. Within the unimodal conditions, users preferred the augmented worktable, but overall, the interindividual usage of all input technologies in the multimodal condition was ranked best. Our findings indicate that the implementation and use of multiple input modalities is valuable, and that it is difficult to predict the usability of individual input modalities for complex systems.

: Border management serves as a crucial control checkpoint for governments to regulate the quality and safety of imported food. In 2020, the ensemble learning prediction model EL V.1 was introduced to Taiwan's border food management. This model primarily assesses the risk of imported food by combining five algorithms to determine whether quality sampling should be performed on imported food at the border. In this study, a second-generation ensemble learning prediction model, EL V.2, was developed based on seven algorithms to enhance the "detection rate of unqualified cases" and improve the robustness of the model. The chi-square test was employed to compare the efficacy of "pre-launch (2019) random sampling inspection" and "post-launch (2020–2022) model prediction sampling inspection". For cases recommended for inspection by the ensemble learning model and subsequently inspected, the unqualified rates were 5.10%, 6.36%, and 4.39% in 2020, 2021, and 2022, respectively, which were significantly higher (p=0.000***) compared with the random sampling rate of 2.09% in 2019. The prediction indices established by the confusion matrix were used to further evaluate the prediction effects of EL V.1 and EL V.2, and the EL V.2 model exhibited superior predictive performance compared with EL V.1, and both models outperformed random sampling.

Fake news detection has become a significant topic based on the fast-spreading and detrimental effects of such news. Many methods based on deep neural networks learn clues from claim content and message propagation structure or temporal information, which have been widely recognized. However, such models (i) ignore the fact that information quality is uneven in propagation, which makes semantic representations unreliable. (ii) Most models do not fully leverage spatial and temporal structure in combination. (iii) Finally, internal decision-making processes and results are non-transparent and unexplained. In this study, we develop a trust-aware evidence reasoning and spatiotemporal feature aggregation model for more interpretable and accurate fake news detection. Specifically, we first design a trust-aware evidence reasoning module to calculate the credibility of posts based on a random walk model to discover high-quality evidence. Next, from the perspective of spatiotemporal structure, we design an evidence-representation module to capture the semantic interactions granularly and enhance the reliable representation of evidence. Finally, a two-layer capsule network is designed to aggregate the implicit bias in evidence while capturing the false portions of source information in a transparent and interpretable manner. Extensive experiments on two benchmark datasets indicate that the proposed model can provide explanations for fake news detection results, as well as can achieve better performance, boosting 3.5% in F1-score on average.

Accurate demand forecasting plays a critical role in most furniture businesses’ operational, tactical, and strategic decisions, as the demand in the furniture business is considered seasonal and becomes more complex in crises. In this work, a neural network model using the Long Short-Term Memory (LSTM) method was developed to forecast the demand for specific product groups. LSTM is a leading deep learning model for time series prediction, particularly seasonal, multi-item, and non-linear situations. The developed model was used to predict the demand based on old data before the Covid-19 pandemic and recent data of the first months of the pandemic as a fast response to the crisis. In addition, a comparison study was conducted between the developed model and the traditional planning inventory used by furniture businesses that provided us with the data. The results showed that the Covid-19 pandemic significantly impacted demand forecasting. Also, the fast response to Covid-19 pandemic has slightly increased the model performance. Finally, the comparison study demonstrated that our model is robust and better than the traditional demand forecasting method. Therefore, the developed model may help the business improve inventory and production planning to create a more flexible supply chain.