Automatic Voice Query Service can extremely reduce the artificial cost, which could improve the response efficiency for users. The automatic speech recognition (ASR) is one of the important component in AVQS. However, many dialect areas in China make the AVQS have to response the multi-accented Mandarin users by single acoustic model in ASR. This problem severely limits the accuracy of ASR for multi-accented speech in the AVQS. In this paper, a new framework for AVQS is proposed to improve the accuracy of response. Firstly, the fusion feature including iVector and filterbank acoustic features is used to train the Transformer-CTC model. Secondly, the transformer-CTC model is used to construct the end-to-end ASR. Finally, key words matching algorithm for AVQS based on fuzzy mathematic theory is proposed to further improve the accuracy of response. The results show that the final accuracy in our proposed framework for AVQS arrives at 91.5%. The proposed framework for AVQS can satisfy the service requirement of different areas in mainland of China. This research has a great significance for exploring the application value of artificial intelligence in the real scene.

The production of butanol, acetone and ethanol by Clostridium acetobutylicum is a biphasic fer-mentation process. In the first phase the carbohydrate substrate is metabolized to acetic and bu-tyric acid, in the following second phase the product spectrum is shifted towards the economi-cally interesting solvents. Here we present a cascade of six continuous stirred tank reactors (CCSTR), which allows performing the time dependent metabolic phases of an ace-tone-butanol-ethanol (ABE) batch fermentation in a spatial domain. Experimental data of steady states under four operating conditions - with variations of the pH in the first bioreactor between 4.3 and 5.6 as well as the total dilution rate between 0.042 1/h and 0.092 1/h - were used to optimize and validate a corresponding mathematical model. Beyond a residence time distribution representation and substrate, biomass and product kinetics this model also includes the differen-tiation of cells between the metabolic states. Model simulations predict a final butanol product concentration of 8.2 g/L and a butanol productivity of 0.75 g/(L h) in the CCSTR operated at a pH in bioreactor 1 of 4.3 and D = 0.092 1/h, while 31 % of the cells are differentiated to the solventogenic state. Aiming at an enrichment of solvent-producing cells, a feedback loop was introduced into the cascade - sending cells from a later state of the process (bioreactor 4) back to an early stage of the process (bioreactor 2). In agreement with the experimental observations, the model accurately predicted an increase of butanol formation rate in bioreactor stages 2 and 3, resulting in an overall butanol productivity of 0.76 g/(L h) for the feedback loop cascade. The here presented CCSTR and the validated model will serve to investigate further ABE fermentation strategies for a controlled metabolic switch.

The potential yield of annual crops is affected by management practices and water and energy availabilities throughout the crop season. The current work aimed to assess the effects of plant population and soil covering on yield components of maize. Field experiments were carried out during 2014-15 and 2015-16 growing seasons at areas grown with oat straw, voluntary plants and bare soil, considering five different plant populations (40,000, 60,000, 80,000, 100,000 and 120,000 plants ha-1) and three sowing dates (15 Sep., 30 Oct., 15 Dec.) for the hybrid P30F53YH in Ponta Grossa, State of Parana, Brazil. Non-impacts of soil covering or plant population on plant height at the flowering phenological stage were observed. Significant effects of soil covering on crop physiological and yield components responses throughout the 2014-15 season were detected. Influence of plant populations on yield components was evidenced, suggesting that from 80,000 plants ha-1 the P30F53YH hybrid performs a compensatory effect among assessed yield components in such a way as to not compromise productivity insofar as plant population increases up to 120,000 plants ha-1. It was noticed a positive trend of yield components and crop final yield as a function of plant density increments.

A photovoltaic-thermal (PVT) collector is a solar-based micro-cogeneration system which generates simultaneously heat and power for buildings. The novelty of this paper is to conduct energy and exergy analysis on PVT collector performance under two different European climate conditions. The performance of the PVT collector is compared to a PV panel. Finally, the PVT design is optimized in terms of thermal and electrical exergy efficiencies. The optimized PVT designs are compared to the PV panel performance as well. The main focus is to find out if the PVT is still competitive with the PV panel electrical output, after maximizing its thermal exergy efficiency. The PVT collector is modelled into Matlab/Simulink to evaluate its performance under varying weather conditions. The PV panel is modelled with the CARNOT toolbox library. The optimization is conducted using Matlab gamultiobj-function based on Non-Dominated Sorting Genetic Algorithm-II (NSGA-II). The results indicated 7.7% higher annual energy production in Strasbourg. However, the exergy analysis revealed a better quality of thermal energy in Tampere with 72.9% higher thermal exergy production. The electrical output of the PVT is higher than from the PV during the summer months. The thermal exergy- driven PVT design is still competitive compared to the PV panel electrical output.

Nowadays, many political campaigns are using social influence (SI) in order to convince voters to support/oppose a specific candidate/party. In election control via SI problem, an attacker tries to find a set of limited influencers to start disseminating a political message in a social network of voters. A voter will change his opinion when he receives and accepts the message. In constructive case, the goal is to maximize the number of votes/winners of a target candidate/party, while in destructive case, the attacker tries to minimize them. Recent works considered the problem in different models and presented some hardness and approximation results. In this work, we consider multi-winner election control through SI on different graph structures and diffusion models, and our goal is to maximize/minimize the number of winners in our target party. We show that the problem is hard to approximate when voters' connections form a graph, and the diffusion model is the linear threshold model. We also prove the same result considering an arborescence under independent cascade model. Moreover, we present a dynamic programming algorithm for the cases that the voting system is a variation of straight-party voting, and voters form a tree.

Acetyleugenol is a phytochemical compound with broad effect against infectious diseases and tumors. Here, we extracted, characterized and elucidated the structure of acetyeugenol, for the first time, from the leaves of Acacia nilotica (L.)―a well-known medicinal plant. The broad antibacterial potential of acetyleugenol was first confirmed against seven bacterial pathogenic isolates with best activity against Proteus sp., Salmonella typhi, Staphylococcu aureus, and Streptococcus pneumonia, which showed similar or better zone of inhibition to that of the control amoxicillin. To further investigate its effect against Mycobacterium tuberculosis, acetyleugenol and its indole and phenyl analogs were subjected to molecular docking experiments against two potential tuberculosis drug targets―MtPknE and MtPknB Ser/Thr protein kinases. The results reveal that all of the analogs have improved docking scores comparing to the acetyleugenol. The indole analogs EUG-1 and EUG-3 were more effective with better docking scores for MtPknE with –11.08 and –10.05 kcal/mol, respectively. Similar results were obtained for the MtPknB. In contrast, only the EUG-2 phenyl analog has given rise to similar docking scores for both targets. This opens the door for further comprehensive studies on these acetyleugenol analogs with in vitro and in vivo experiments to validate and get more insights into their mechanisms of action.

Diabetes mellitus (DM) is one of the major risk factors for COVID-19 complications as it is one of the chronic immune-compromising conditions especially if patients have uncontrolled diabetes, poor HbA1c &/or irregular blood glucose levels. Diabetic patient’s mortality rates with COVID-19 are higher than cardiovascular or cancer patients. Recently Bacillus Calmette–Guérin (BCG) has shown successful results in reversing diabetes in both rats and clinical trials based on different mechanisms from aerobic glycolysis to Beta cells regeneration. BCG is a multi-face vaccine that has been used extensively in protection from TB and leprosy and has been repositioned for treatment of bladder cancer, diabetes & multiple sclerosis. Recently, the COVID-19 epidemiological study confirmed that universal BCG vaccination reduced morbidity and mortality in certain geographical areas. Countries without universal policies of BCG vaccination (Italy, Nederland, USA) have been more severely affected compared to countries with universal and long-standing BCG policies that have shown low numbers of reported COVID-19 cases. Some countries have started clinical trials that included a single dose BCG vaccine as prophylaxis from COVID-19 or an attempt to minimize its side effects. This proposed research aims to use BCG vaccine as a double-edged weapon countering both COVID-19 & diabetes, not only as protection but also as therapeutic vaccination. The work includes a case study of regenerated pancreatic beta cells based on improved C-peptide & PCPRI laboratory findings after BCG vaccination for a 9 years’ patient. The patient was re-vaccinated based on a negative tuberculin test & no scar at the site of injection of the 1st BCG vaccination at birth. Furthermore, the authors in the present article described a prospective BCG multi-dose clinical study in full details that they will apply in case of acceptance of their submitted grant & the ethical committee approval. The aim of the clinical study is to check if double dose BCG (4 weeks apart) will show a significant difference in the protection of health care professionals in Egypt. The authors suggest and invite the scientific community to take into consideration the concept of direct BCG re-vaccination (after 4 weeks) because of the reported gene expressions & exaggerated innate immunity consequently. As the diabetic MODY-5 patient (mutation of HNF1B, Val2Leu) was on low dose Riomet® while eliminating insulin gradually, a simple analytical method for metformin assay was recommended to ensure its concentration before use as it is not approved yet by the Egyptian QC labs.

The need to understand the process by which particles, including solar wind and coronal ions as well as pickup ions, are accelerated to high energies (ultimately to become anomalous cosmic rays) motivate a multi-fluid shock wave model which includes kinetic effects (e.g. ion acceleration) in an electromagnetically self-consistent framework. Particle reflection at the cross-shock potential leads to ion acceleration in the motional electric field and thus anisotropic heating and pressure in the shock layer, with important consequences for the multi-fluid dynamics. This motivates development of a multi-fluid model of solar wind electrons and ions treated as fluid, coupled self-consistently with a small population of kinetically treated ions (e.g. pickup ions.) Consideration of both the time dependent and steady state regimes, indicate that such a multi-fluid approach is necessary for resolving the, Debye scale, particle reflecting cross-shock potential and subsequent dynamics. To study charge separation effects in narrow, supersonic wave layers we consider a reduction of the system to the steady state for cold ions and hot electrons and find two types of solitary waves inherent to the reduced two-fluid system in this limiting case.

Industry 4.0 is having a great impact in all smart efforts. This is not a single product, but is composed of several technologies, being one of them Industrial Internet of Things (IIoT). Currently, there are very varied implementation options offered by several companies, and this imposes a new challenge to companies that want to implement IoT in their processes. This challenge suggests to use multi-criteria analysis to make a repeatable and justified decision, requiring a set of alternatives and criteria. This paper proposes a new methodology and comprehensive criteria to help organizations to take an educated decision by applying multi-criteria analysis. Here, we suggest a new original use of PROMETHEE-II with full example from weight calculation up to IoT platform selection, showing this methodology as an effective study for other organizations interested to select an IoT platform. The criteria proposed outstands from previous work by including not only technical aspects, but economic and social criteria, providing a full view of the problem analyzed. A case of study was used to prove this proposed methodology.

Internet of things (IoT) is a network of smart things. This indicates the ability of these physical things to transfer information with other physical things. IoT has introduced various services and daily human life depends on its reliable and accessible operation. The characteristics of these networks, such as topology dynamicity and energy constraint, challenges the routing problem in these networks. Previous routing methods could not achieve the required performance in this type of network. Therefore, developers of this network designed and developed specific methods in order to satisfy the requirements of these networks. One of the routing methods is utilization of multi-path protocols which send data to its destination using routs with separate links. One of such protocols is AOMDV routing protocol. AOMDV protocol is a multi-path protocol which uses multiple different paths for sending information in order to maintain the network traffic balance, manage and control node energy, decrease latency, etc. In this paper, this method is improved using gray system theory which chooses the best paths used for separate routes to send packets. To do this, AOMDV packet format is altered and some fields are added to it so that energy criteria, link expiration time, and signal to noise ratio can also be considered while selecting the best route. The proposed method named GSTMPR-IoT is introduced which chooses the routs with highest rank for concurrent transmission of data, using a specific routine based on the gray system theory. In order to evaluate and report the results, the proposed GSTMPR-IoT method is compared to the EECRP and AOMDV approaches with regard to throughput, packet delivery rate, end to end delay, average residual energy, and network lifetime. The results demonstrate the superior performance of the proposed GSTMPR-IoT compared to the EECRP and AOMDV approaches.

The immunomodulatory behaviour of the CCR6/CCL20 axis in multi -system pathophysiology and molecular signalling was investigated at two clinically significant time points, using a Ccr6 - deficient mouse model of spontaneous colitis. Four groups of mice, (C57BL/6J, Ccr6^-/- of C57BL/6J, Winnie x Ccr6 ^-/- and Winnie) were utilized and (I) colonic clinical parameters (2) histology of colon, spleen, kidney and liver (3) T and B lymphocyte distribution in the spleen and MLN by flowcytometry (5) colonic CCL20, phosphorylated PI3K and phosphorylated Akt expression by immunohistochemistry and (6) colonic cytokine expression by RT-PCR were evaluated. CCR6 deficiency was shown to attenuate inflammation in the spleen, liver and gut while renal histology remained unaffected. Marked focal lobular inflammation with reactive nuclear features were observed in hepatocytes and a significant neutrophil infiltration in red pulp with extra medullary hemopoiesis in the spleen existed in Winnie. These changes were considerably reduced in Winnie x Ccr6^-/- with elevated goblet cell numbers and mucus production in the colonic epithelium. Results indicate that Ccr6- deficiency in the colitis model contributes towards resolution of disease. Our findings demonstrate an intricate networking role for CCR6 in immune activation, which is downregulated by Ccr6 deficiency, and could provide newer clinical therapies in colitis.

In this paper we show how to efficiently implement parallel discrete simulations on Multi-Core and GPU architectures through a real example of application: a cellular automata model of laser dynamics. We describe the techniques employed to build and optimize the implementations using OpenMP and CUDA frameworks. We have evaluated the performance on two different hardware platforms that represent different target market segments: high-end platforms for scientific computing, using an Intel Xeon Platinum 8259CL server with48cores and also an NVIDIA Tesla6V100 GPU, both running on Amazon Web Server (AWS) Cloud, and on a consumer-oriented platform, using an Intel Core i9 9900k CPU and an NVIDIA GeForce GTX 1050 TI GPU. Performance results are compared and analysed in detail. We show that excellent performance and scalability can be obtained in both platforms, and we extract some important issues that imply a performance degradation for them. We also found that current Multi-Core CPUs with large core numbers can bring a performance very near to that of GPUs, even similar in some cases.

This study presents a path planning method for a mobile robot to be effectively operated through a multi-objective decision-making problem. Specifically, the proposed Fuzzy analytic hierarchy process (FAHP) determines an optimal position as a sub-goal within the multi-objective boundary. The key features of the proposed FAHP are evaluating the candidates according to the fuzzified relative importance among objectives to select an optimal solution. In order to incorporate FAHP into path planning, an AHP framework is defined, which includes the highest level (goal), middle level (objectives), and the lowest level (alternatives). The distance to the target, robot’s rotation, and safety against collision between obstacles are considered as objective functions. Comparative results obtained from the artificial potential field and AHP/FAHP simulations show that FAHP is much preferable for the mobile robot’s path planning than typical AHP.

Two desirable functions of healthy soil are nutrient cycling and pest suppression. We review recent literature on the contributions of soil microarthropods to soil health through their intersecting roles in decomposition and nutrient cycling and direct and indirect suppression of plant pests. Microarthropods can impact soil and plant health directly by feeding on pest organisms or serving as alternate prey for larger predatory arthropods, and indirectly, by mediating the ability of crop plants to resist or tolerate insect pests and diseases through interactions with the decomposition food web in support of plant nutrition. Soil fauna, including microarthropods, are key regulators of decomposition at local scales but their role at larger scales is unresolved. Future research priorities include the incorporation of multi-channel omnivory into food web modeling and understanding the vulnerability of our soil carbon to increased global temperatures.

The growth of demand for metallic minerals has faced with the need for new techniques and improving technologies for all mine life cycle operations. Nowadays, the exploitation of old tailings and mine wastes facilities could represent a solution to this demand, with economic and environmental advantages. W-Sn Panasqueira Mine has been operating for more than 100 years. Its first processing plant “Rio” was located near Zêrere river being the mineral processing residues deposited on the top hillside on the margin of this river in Cabeço do Pião tailings dam. The lack of maintenance and monitoring of this enormous structure in the last twenty years represents high risks to the environment and population of the surrounding region. The re-mining of the tailings by hydrometallurgical methods was considered, in order to satisfy these two enounced conditions - metals demand and environmental risk, aiming for the sale of the metal to pay the environmental intervention. Field samples campaign allowed collecting data and results from laboratory tests driving to use regression optimization. The re-mining solution was studied, taking into account the technical, economic, social, and environmental aspects.

The combination of new-generation information technology and manufacturing technology has resulted in major and profound impact on future development paradigm of manufacturing. It is challenging for existing methods to conduct a multidimensional trend exploration related to machine tool domain, which is the basis of virtually everything in manufacturing. In this paper, we proposed an integrating approach framework combined topic models, bibliometric, trend analysis and patent analysis to mine insightful information about future development from multi-source data related to machine tool, such as papers, grants, patents and news. Specifically, papers and grants provided two different perspectives to explore the current focuses and future trends in machine tool research. Furthermore, the future technology development of machine tool was investigated through patents analysis. Finally, news related to machine tool industry in recent years was analyzed to analyze future machine tool business mode. The integration of the above various analytical methods and the multi-dimensional mining of literatures enabled the analysis of the future development of machine tool domain systematically from multi-perspectives which include research, technology development and industry. The conclusions obtained in this paper is beneficial to different communities of machine tool in terms of determining the research directions for researchers, identifying industry opportunities for corporations and developing reasonable industry policy for policy makers.

For the task planning problem of the command and control architecture, the existing algorithms have problems such as low efficiency and poor re-planning quality under abnormal conditions. Based on the requirements of the current accusation architecture, this paper constructs a distributed command and control architecture model based on multi-agents, which makes use of the superiority of multi-agents in dealing with complex tasks. The concept of MultiAgents-HTN is proposed under the framework. The original hierarchical task network planning algorithm is optimized, the multi-agent collaboration framework is redefined, and the coordination mechanism of local conflict is designed. Taking the classical resource scheduling problem as the experimental background, the comparison between the proposed algorithm and the classical HTN algorithm is carried out. The experimental results show that the proposed algorithm has higher quality and higher efficiency than the existing algorithm, and the space anomaly is heavy during processing. The planning is more efficient, and the time is more complicated and superior in dealing with the same problem, with good convergence and adaptability. The conclusion proves that the distributed command and control architecture proposed in this paper has high practicability in related fields and can solve the problem of distributed command and control architecture in multi-agent environment.

In this paper, we propose a novel approach of video skimming by exploiting the fusion of video temporal information and keyword information representation extracted from multi-model video information including audio, text and visual indices. In addition, we introduce the brand-safe filtering and sentiment analysis in order to only reserve the user-friendly content in the video skim. In the experiment by using the videos from YouTube-8M dataset, we have proved that the semantic conservation in the video skim from the proposed approach highly outperforms the approaches by only partial information of the video in conserving the semantic content of the video.

The electrocardiogram (ECG) was the first biomedical signal where digital signal processing techniques were extensively applied. By its own nature, the ECG is typically a sparse signal, composed of regular activations (the QRS complexes and other waveforms, like the P and T waves) and periods of inactivity (corresponding to isoelectric intervals, like the PQ or ST segments), plus noise and interferences. In this work, we describe an efficient method to construct an overcomplete and multi-scale dictionary for sparse ECG representation using waveforms recorded from real-world patients. Unlike most existing methods (which require multiple alternative iterations of the dictionary learning and sparse representation stages), the proposed approach learns the dictionary first, and then applies an efficient sparse inference algorithm to model the signal using the learnt dictionary. As a result, our method is much more efficient from a computational point of view than other existing methods, thus becoming amenable to deal with long recordings from multiple patients. Regarding the dictionary construction, we locate first all the QRS complexes in the training database, then we compute a single average waveform per patient, and finally we select the most representative waveforms (using a correlation-based approach) as the basic atoms that will be resampled to construct the multi-scale dictionary. Simulations on real-world records from Physionet's PTB database show the good performance of the proposed approach.

Tissue engineering is an emerging field of research due to its growing impact on the regeneration of tissue injured from damage and organ failure. Small hydrogel microspheres containing regenerative cells and stimulators can be used as building blocks to regenerate tissue by injecting them into damaged areas; a microneedle is used to minimize the incision and any associated infections. To address the need for production of a large number of microspheres for use in tissue engineering, we developed a 3D coaxial multi-nozzle flow-focusing device fabricated via a simple 3D-printing method that demonstrates a high rate of microsphere production per nozzle. This device has six coaxial microscale nozzles that produce hydrogel microspheres. Two individual parts—the inlet and nozzle—can be made separately by a 3D printer and bonded together using uncured photo-curable resin as glue. The dimensions of the microspheres are between 100 μm and 1200 μm. They are produced by adjusting the flow rate ratio between the dispersed and continuous media. A flow rate ratio of 180 demonstrated the highest microsphere production rate of 2.12e+5 microspheres per sec (0.25 mL/min). This microsphere production rate per nozzle is four times higher than that of currently available devices.

The paper analyses how between 1956 and 2009 the agrarian metabolism of the Barcelona Metropolitan Region (BMR) has become less functional, losing circularity in biomass flows and in relationship to its landscape. We do so by adopting a Multi-EROI and flow-fund (MuSIASEM) analyses and its nexus with landscape functional structure. The study of agricultural flows of Final Produce, Biomass Reused and External Inputs is integrated with that of land use, livestock, power capacity and population changes between 1956 (at the beginning of the Green Revolution) and 2009 (fully industrialized agriculture). A multi-scale analysis is conducted at the landscape scale (seven districts within the Barcelona metropolitan region) as well as for the functions deployed, within an agroecosystem, by the mutual interactions between its funds (land-uses, livestock and farming population). A complex nexus between land, livestock, dietary patterns and energy needs is shown; we conclude that from the perspective of the circular bioeconomy the agrarian sector has gone worse hand in hand with the landscape functional structure. Therefore, a novel perspective in landscape agroecology is opened

This study shows the application of self-organizing maps (SOMs) to probabilistic forecasts of wind power generation and ramps in Japan. SOMs are applied to atmospheric variables obtained from atmospheric reanalysis over the region, thus deriving classified weather patterns (WPs). Probabilistic relationships are established between the synoptic-scale atmospheric variables over East Japan and the generation of regionally integrated wind power in East Japan. Medium-range probabilistic wind power predictions are derived by SOM, as analog ensembles based on the WPs of the multi-center ensemble forecasts. As this analog approach handles stochastic uncertainties effectively, probabilistic wind power forecasts are rapidly generated from a very large number of forecast ensembles. The use of a multi-model ensemble provides better results than a one-forecast model. The hybrid ensemble forecasts further improve the probabilistic predictability skill of wind power generation, as compared with non-hybrid methods. It is expected that long-term wind forecasts will provide better guidance to transmission grid operators. The advantage of this method is that it can include an interpretative analysis of meteorological factors for variations in renewable energy.

Physical frailty and sarcopenia (PF&S) are hallmarks of aging that share a common pathogenic background. Perturbations in protein/amino acid metabolism may play a role in the development of PF&S. In this preliminary study, 68 community-dwellers aged 70 years and older, 38 with PF&S and 30 non-sarcopenic, non-frail controls (nonPF&S), were enrolled. A panel of 37 serum amino acids and derivatives was assayed by UPLC-MS. Partial Least Squares Discriminant Analysis (PLS-DA) was used to characterize the amino acid profile of PF&S. The optimal complexity of the PLS-DA model was found to be three latent variables. The proportion of correct classification was 76.6 ± 3.9% (75.1 ± 4.6% for enrollees with PF&S; 78.5 ± 6.0% for controls). Older adults with PF&S were characterized by higher levels of asparagine, aspartic acid, citrulline, ethanolamine, glutamic acid, sarcosine, and taurine. The profile of nonPF&S individuals was defined by higher levels of α-aminobutyric acid and methionine. Distinct profiles of circulating amino acids and derivatives characterize older individuals with PF&S. The dissection of these patterns may provide novel insights into the role played by protein/amino acid perturbations in the disabling cascade and possible new targets for interventions.

The activity of prioritizing software requirements should be done as efficiently as possible. Selecting the most stable requirements for the most important customers for the development company can be a positive factor when we consider that the available resource does not always encompass the implementation of all requirements. Quantitative methods for reaching software prioritization in releases are many in the field of Search-Based Software Engineering (SBSE). However, we show that it is possible to use qualitative Verbal Decision Analysis (VDA) methods to solve this same type of problem. Moreover, we will use the ZAPROS III-i methods to prioritize requirements considering the opinion of the decision-maker, who will participate in this process. Finally, the results obtained in the VDA structured methods were quite satisfactory when compared to the methods using SBSE. A comparison of results between quantitative and qualitative methods will be made and discussed later.

Landing is one of the most dangerous tasks in all the operations on the aircraft carrier, and the landing safety is very important to the pilot and the flight deck operation. The problem of landing path selection is studied in this paper as there several candidates corresponding to different situations. A fuzzy path selection strategy is proposed to solve the problem considering the fuzziness of environmental information and human judgment, and the goal is to provide the pilot with more reasonable decision. The strategy is based on Fuzzy Multi-attribute Group Decision Making (FMAGDM), which has been widely used in industry. Firstly, the background of the path selection problem is given. Then the essential elements of the problem are abstracted to build the conceptual model. A group decision-making method is applied to denote the preference of each decision maker for each alternative route, and the optimal landing path under the current environment is determined taking into account the knowledge and the weight of both decision makers. Experimental studies under different setups, i.e., different environments, are carried out. The results demonstrate that the proposed path selection strategy is validated in different environments, and the optimal landing paths corresponding to different environments can be determined.

The eddy current loss should be optimized to be as less as possible for the stability of permanent magnet in high speed permanent magnet synchronous motor (HSPMSM) rotor and ensure the high efficiency and low temperature of the motor. This paper analyzes the eddy current distribution in rotor, with consideration of the conflict of the thickness of sleeve and diameter of the rotor, calculating the eddy current loss (ECL) and the thermal distribution via Separation of variables method for solving Maxwell's equations with analytical hieratical model of ECL constructed. The optimization result of ECL of the HSPMSM whose power and rated speed is 30kw 48000r/min can be got by multi-objective optimization method, combined weighting coefficient method and traversal algorithm based on chaotic local search particle swarm optimization (CLSPSO), utilizing ECL analytical model and other analytical constraints. Related experiment and measurement has been implemented with new approach of loss separation.

The planning of rural land depends to a large extent on the agrarian and forestry development model agreed upon for it. Within the framework of agrarian multifunctionality, to evaluate different development alternatives and their effect on territorial planning, its multiple dimensions must be considered as well as the interests and needs of the stakeholders. This work thus addresses the comprehensive evaluation of development alternatives for the rural area of a municipality of the Basque Country (Spain) through the implementation of a multi-criteria evaluation method with social participation. The results show, firstly, the existence of trade-offs between the functions to be promoted in the various development models evaluated. Secondly, the results indicate the need to reverse recent trends in the sector and the convenience of moving towards an agro-livestock model of greater ecological and local character, independently of the followed forest model. In fact, the evaluation carried out also reveals, thirdly, the existence of more or less latent conflicts, in particular with regards to forest policy.

This paper presents a multi-swarm PSO algorithm for the Quadratic Assignment Problem (QAP) implemented on the OpenCL platform. Our work was motivated by results of time efficiency tests performed for single-swarm algorithm implementation that showed clearly that the benefits of a parallel execution platform can be fully exploited provided the processed population is large. The described algorithm can be executed in two modes: with independent swarms or with migration. We discuss the algorithm construction as well as we report results of tests performed on several problem instances from the QAPLIB library. During the experiments the algorithm was configured to process large populations. This allowed us to collect statistical data related to values of goal function reached by individual particles. We use them to demonstrate on two test cases that although single particles seem to behave chaotically during the optimization process, when the whole population is analyzed, the probability that a particle will select a near-optimal solution grows.

Surface damage on honeycomb aircraft panels is often measured manually, and is therefore subject to variation based on inspection personnel. Eddy current testing (ECT) is sensitive to variations in probe-to-specimen spacing, or lift-off, and is thus promising for high resolution profiling of surface damage on aluminum panels. Lower frequency testing also allows inspection through the face sheet, an advantage over optical 3D scanning methods. This paper presents results from the ECT inspection of surface damage on an approximately flat aluminum honeycomb aircraft panel, and compares the measurements to those taken using optical 3D scanning technology. An ECT C-Scan of the dented panel surface was obtained by attaching the probe to a robotic scanning apparatus. Data was taken simultaneously at four frequencies of 25, 100, 400 and 1,600 kHz. A reference surface was then defined that approximated the original, undamaged panel surface, which also compensated for the effects of specimen tilt and thermal drift within the ECT instrument. Data was converted to lift-off using height calibration curves developed for each probe frequency. A damage region of 22,550 mm² area with dents ranging in depth from 0.13-1.01 mm was analyzed. The method was accurate at 1,600 kHz to within 0.05 mm (2s) when compared with 231 measurements taken via optical 3D scanning. Testing at 25 kHz revealed a 3.2 mm cell size within the honeycomb core, which was confirmed via destructive evaluation. As a result, ECT demonstrates potential for implementation as a method for rapid in-field aircraft panel surface damage assessment.

Satellite scene classification is challenging because of the high variability inherent in satellite data. Although rapid progress in remote sensing techniques has been witnessed in recent years, the resolution of the available satellite images remains limited compared with the general images acquired using a common camera. On the other hand, a satellite image usually has a greater number of spectral bands than a general image, thereby permitting the multi-spectral analysis of different land materials and promoting low-resolution satellite scene recognition. This study advocates multi-spectral analysis and explores the middle-level statistics of spectral information for satellite scene representation instead of using spatial analysis. This approach is widely utilized in general image and natural scene classification and achieved promising recognition performance for different applications. The proposed multi-spectral analysis firstly learns the multi-spectral prototypes (codebook) for representing any pixel-wise spectral data, and then based on the learned codebook, a sparse coded spectral vector can be obtained with machine learning techniques. Furthermore, in order to combine the set of coded spectral vectors in a satellite scene image, we propose a hybrid aggregation (pooling) approach, instead of conventional averaging and max pooling, which includes the benefits of the two existing methods but avoids extremely noisy coded values. Experiments on three satellite datasets validated that the performance of our proposed approach is much more accurate than even the deep learning framework for spatial analysis.

This paper deals with a generic methodology to evaluate the magnetic parameters of contactless power transfer systems. Neumann's integral has been used to create a matrix method that can model the magnetics of single coils (circle, square, rectangle). The principle of superposition has been utilised to extend the theory to multi-coil geometries such as double circular, double rectangle and double rectangle quadrature assuming linearity of magnetics. Numerical and experimental validation has been performed to validate the analytical models developed. A rigorous application of the analysis has been carried out to study misalignment and hence the efficacy of various geometries to misalignment tolerance. Comparison of single-coil and multi-coil shapes considering coupling variation with misalignment, power transferred and maximum efficiency is carried out.

Face verification for unrestricted faces in the wild is a challenging task. This paper proposes a method based on two deep convolutional neural networks(CNN) for face verification. In this work, we explore to use identification signal to supervise one CNN and the combination of semi-verification and identification to train the other one. In order to estimate semi-verification loss at a low computation cost, a circle, which is composed of all faces, is used for selecting face pairs from pairwise samples. In the process of face normalization, we propose to use different landmarks of faces to solve the problems caused by poses. And the final face representation is formed by the concatenating feature of each deep CNN after PCA reduction. What's more, each feature is a combination of multi-scale representations through making use of auxiliary classifiers. For the final verification, we only adopt the face representation of one region and one resolution of a face jointing Joint Bayesian classifier. Experiments show that our method can extract effective face representation with a small training dataset and our algorithm achieves 99.71% verification accuracy on LFW dataset.

The study of different forms of preconditioners for solving a system of nonlinear equations, by using Newton’s method, is presented. The preconditioners provide numerical stability and rapid convergence with reasonable computation cost, whenever chosen accurately. Different families of iterative methods can be constructed by using a different kind of preconditioners. The multi-step iterative method consists of a base method and multi-step part. The convergence order of base method is quadratic and each multi-step add an additive factor of one in the previously achieved convergence order. Hence the convergence of order of an m-step iterative method is m + 1. Numerical simulations confirm the claimed convergence order by calculating the computational order of convergence. Finally, the numerical results clearly show the benefit of preconditioning for solving system of nonlinear equations.

We present information theoretic search strategies for single and multi-robot teams to localize the source of biochemical contaminants in turbulent flows. The robots synthesize the information provided by sporadic and intermittent sensor readings to optimize their exploration strategy. By leveraging the spatio-temporal sensing capabilities of a mobile sensing network, our strategies result in control actions that maximize the information gained by the team while optimizing the time spent localizing the position of the biochemical source. By leveraging the team's ability to obtain simultaneous measurements at different locations, we show how a multi-robot team is able to speed up the search process resulting in a collaborative information theoretic search strategy. We validate our proposed strategies in both simulations and experiments.

Scene classification plays an important role in the intelligent processing of high-resolution satellite (HRS) remotely sensed image. In HRS image classification, multiple features, e.g. shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS image. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structure across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.

Predicting electricity prices is a very important issue in modern society, because the associated decision process under uncertainty requires accurate forecasts for the economic agents involved. In this paper, we apply the decision tree extension of Random Forests to the prediction of electricity prices in Spain, but with the novelty of modeling prices jointly with demand, with the purpose of achieving greater accuracy than with univariate response Random Forests, particularly in price prediction, as well as understanding the effect of the input variables (lagged values of price and demand, current production levels of available energy sources) on the joint of the two outputs. The results are very encouraging, providing significant increase in price prediction accuracy. Also, interesting methodological challenges appear as far as the appropriate choice of the relative weights of price and demand in the joint modeling is concerned and a new procedure to provide the importance variable ranking is proposed. The partykit (package of R software) library allowing for multivariate Random Forests has been used.

Background/Objectives: Interest in diversity is growing worldwide. Today, understanding and social acceptance of diverse people is becoming increasingly important. Therefore, in this study, we aimed to clarify the relationship between gray matter volume (GMV), which is thought to reflect brain health, and understanding of diversity (gender, LGBTQ, and race). Methods: GMV was the value of the Gray Matter Brain Healthcare Quotient (GM-BHQ) based on MRI image analysis. Meanwhile, diversity was calculated from the answers to the psychological questions included in the World Values Survey Wave 7 (WVS7). Results: As a result of the analysis, in the group of participants with the highest understanding of diversity (PHUD. n = 11), not only the GMV at the whole brain level but also the GMV of the central executive network (CEN) and saliency network (SN) were shown to be significantly higher than the theoretical value estimated from sex, age, and BMI at the 5% level. In addition, the GMV of the default mode network (DMN) was also shown to be higher than the theoretical value at the 10% level. Meanwhile, in the group of others (n = 10), there was no significant difference from the theoretical value. These differences between PHUD and others were also observed when comparing the two. Conclusions: These results suggest that understanding diversity requires a healthy brain, centered on three networks that govern rational judgment, emotion regulation, other-awareness, self-awareness, and the valuing of actions. This is the first study to show that brain structure is related to understanding and acceptance of diverse people.

Detecting people and technical objects in various situations, such as natural disasters and warfare, is critical to search and rescue operations and the safety of the civilian. A fast and accurate detection for people and equipment can significantly increase the effectiveness of search and rescue missions and provide timely assistance to people. Computer vision and deep learning technologies play a key role in detecting the required objects due to their ability to analyse big volumes of visual data in real time. The performance of the neural networks such as YOLOv4-v8, Faster R-CNN, SSD, and EfficientDet has been analysed using COCO2017, SARD, SeaDronesSee, and VisDrone2019 datasets. Main criteria for comparison were mAP, Precision, Recall, F1-Score, and the ability of the neural network to work in real time. The most important metrics for evaluating the efficiency and performance of models for a given task are accuracy (mAP), F1-Score, and processing speed (FPS). These metrics allow us to evaluate both the accuracy of object recognition and the ability to use the models in real-world environments where high processing speed is important. Although different neural networks perform better on certain types of metrics, YOLO outperforms them on all metrics, showing the best results of mAP-0.88, F1 - 0.88, and FPS - 48, so the focus was on these models.

Traditional chemical research often relies on trial-and-error synthesis and characterization methods. Now, modern machine learning (ML) offers data-driven approaches for predicting properties, like water solubility, directly from chemical structure. But with various data representation schemes for molecular structure and model approaches to select from, it can be difficult for non-experts to determine best practices for utilizing ML. To clarify this landscape of choices, this study uses the ESOL molecular solubility dataset to compare the performance of a selection of different models on different data representations. First, we compare three classical regression methods (linear, ridge, LASSO) on three common data representations (RDKit fingerprint, Morgan fingerprint, intuition-selected molecular features). Then, we demonstrate how two distinct deep learning approaches (multilayer perceptron, graph convolution) can achieve accurate predictions even when prior intuition about feature-property correlations are absent. Finally, we outline a modern attention-based approach, inspired by successes in language modeling and fine-tuned by Bayesian optimization, to achieve a prediction methodology that is more general and performant than the previous approaches. This attention-based approach operates directly on the common SMILES string molecular representation, without requiring as many model parameters as other deep learning approaches or preprocessing into a representative fingerprint or vector of intuitively selected features. In short, when selected molecular features are known to likely correlate with a feature of interest, it may be possible to achieve good predictive modeling without turning to massive deep learning approaches. When particular features are not known a priori, graph approaches are a common solution, and we further demonstrate how a modern hyperparameter optimized attention approach can perform even better.

Parkinson's disease (PD) is a prevalent neurodegenerative disorder marked by progressive motor and non-motor symptoms. Traditional treatments primarily manage symptoms rather than address the disease's root causes, highlighting the need for novel therapeutic approaches. Natural phytochemicals, particularly phenolic compounds, have shown potential due to their neuroprotective effects. Antiaris africana is commonly used in traditional African medicines to treat several illnesses including neurodegenerative disorders. However, its neuroprotective mechanism remains elusive. This study aimed to employ network pharmacology-based prediction to identify key targets and pathways through which A. africana exerts its effects on PD. Targets associated with Parkinsonism were identified and analyzed for interaction with compounds from A. africana. Protein-protein interaction (PPI) networks were constructed, and key hub genes were identified using Cytoscape. Functional enrichment analysis highlighted significant pathways involved in PD. Molecular docking was performed to evaluate the binding affinities of the compounds to PD-related targets. Our results reveal that compounds such as caffeic acid, chlorogenic acid, catechin, ellagic acid, epigallocatechin, gallic acid, isoquercitrin, kaempferol, quercetin, quercitrin, and rutin interact with key proteins implicated in PD, including APP, EGFR, CASP3, ESR1, AKT1, and MAPK8. Notably, rutin exhibited strong binding affinity for EGFR, and chlorogenic acid showed potential in modulating the APP pathway. These findings suggest that phenolic compounds in A. africana may exert neuroprotective effects through multiple molecular mechanisms, offering potential therapeutic strategies for PD. This study underscores the importance of integrating molecular docking and network pharmacology to elucidate the therapeutic potential of natural compounds in complex diseases like Parkinson's disease

Tackling the intricacies of network dismantling in complex systems poses significant challenges. This task has relevance across various practical domains, yet traditional approaches focus primarily on singular metrics, such as the number of nodes in the Giant Connected Component (GCC) or average pairwise connectivity. In contrast, we propose a unique metric that concurrently targets nodes with the highest degree and reduces the GCC size. Given the NP-hard nature of optimizing this metric, we introduce MaxShot, an innovative end-to-end solution that leverages graph representation learning and reinforcement learning. Through comprehensive evaluations on both synthetic and real-world datasets, our method consistently outperforms leading benchmarks in accuracy and efficiency. These results highlight MaxShot’s potential as a superior approach to address the network dismantling problem effectively.

Textile heaters are made of knitted conductive yarns integrated into the fabric, making them stretchable, washable, breathable, and suitable for close-to-skin wear. The non-zero resistance in the lead wires, however, causes non-uniform power distribution, presenting a design challenge. To address this, the electrical performance of the heaters is modelled as an n-ladder resistor network. Using the finite difference method, simple, closed-form expressions are derived for networks with the power source connected to input terminals A1B1 and A1Bn, respectively. The results are exact and are used to derive approximations and design criteria. Solutions for ladder networks as presented in this paper apply to a wider class of physical problems, such as irrigation systems, transformer windings, and cooling fins.

Nowadays and in the web era, there is a huge amount of information available due to automatically and easily generated websites. This information has many shared opinions from people toward different entities. Over the last decade there was a significant huge interest to determine the sentiment of any type of text especially on twitter. Twitter provides unstructured text that has many misspelling words, slang words and shortened forms. More complexity and challenges have come to the sentiment analysis researchers in the context of twitter. In this paper we conducted a literature review of the recent research of Twitter sentiment analysis on different languages. After that, we investigated challenges related to twitter sentiment analysis of Arabic language. Then we found a new solution for Sentiment analysis of Arabic twitter. So, we designed the main component of the proposed system and presented its algorithm. Finally, we performed many experiments under two kinds of datasets. The system achieves highest accuracy and shows many important results..

Human hair is characterised by significant variability, determined by genetic and macromolecular factors. Whilst European hair type has been a focus of extensive research, Afro-textured hair care faces challenges created by insufficient knowledge of its unique properties. Applications of hair care products that are incompatible with Afro-textured hair frequently have detrimental effects on the scalp, including alopecia. This highlights the need for scientific and clinical partnerships to bridge the gap between research and hair care practices and address challenges related to Afro-textured hair. In this review, we performed data mining of the existing literature and in-silico network analysis of the biomarkers relevant to Afro-textured hair. The approaches to hair maintenance are highlighted in the context of hair anatomy and growth cycles, organisation of keratins, surface lipids and chemical bonds. We discuss a range of biomarkers affecting hair fibre's shape and mechanical strength, with the gene interactive network pointing to the hierarchical organisation of important traits, notably hair shaft diameter, keratinisation, hair follicle development and patterning. We propose a better understanding of the genetic traits, molecular structure and biomechanics of Afro-textured hair is required to initiate more effective hair care solutions that would benefit the wider population.

Background: Inflammatory plays critical role in myocardial infarction as a critical process in the development of heart failure, involving the development of cardiac fibrosis. Colchicine is a well-established anti-inflammatory drug, but its scientific application in controlling post-acute myocardial infarction remodelling processes has not been established. IL-10, a key cytokine in modulating inflammatory responses, underscoring its potential as a crucial therapeutic target of colchicine. Objectives: To explore the protective role of IL-10 modulated by colchicine in myocardial healing and repair following AMI, particularly cardiac fibrosis. Materials and Methods: The predicted protein of Colchicine was assessed using WAY2DRUG PASS as probability active value. Proteins associated with Colchicine, cardiac fibrosis and acute myocardial infarction were analysed with DisGeNET and Open Target databases. Analysis and visualization of protein-protein interactions were conducted using STRING and Cytoscape. A 3T3 cell line treated with CoCl2 was used to mimic hypoxic. HIF-1α and IL-10 expression were measured by flow cytometry and analysed using a one-way ANOVA test. This observational clinical trial examined acute myocardial infarction patients undergoing immediate and delayed primary percutaneous coronary interventions. Subjects were randomized into control groups receiving placebo and intervention groups treated with Colchicine. Assessments occurred at 24 hours and five days after the intervention. IL-10 expression in the clinical trial was measured by ELISA assay and analysed using a T-test. Results: Colchicine demonstrates promising bioactivity in treating acute myocardial infarction, with notable probability active values as a tubulin antagonist (0.744), beta-tubulin antagonist (0.673), and NOS2 inhibitor (0.529). Its primary action targets IL-10, with protein-protein interactions analysis indicating interactions between IL-10 and key inflammatory mediators—IL-1β, IFN-γ, CCL2, TNF, and TGF-β1—during acute myocardial infarction and cardiac fibrosis. Hypoxic conditions in the CoCl2-induced 3T3 cell model show significantly elevated HIF-1α compared to controls (p<0.0001). Acute myocardial infarction patients were predominantly from high-risk populations, with notable smoking prevalence in the early Colchicine group (72%). Hypertension prevalence was significantly different, with 67.31% in the late Colchicine group compared to 38% in the early Colchicine and 39.80% in the early placebo groups. Both early and late Colchicine groups exhibited a higher prevalence of diabetes mellitus. Colchicine use, either early or late, did not significantly alter major adverse cardiovascular events incidence compared to placebo (p>0.05). However, Colchicine significantly increased IL-10 expression in CoCl2-treated cells (p<0.0001) and in acute myocardial infarction patients within five days (p<0.05). Conclusions: Colchicine may bolster the anti-inflammatory response post-myocardial infarction by activating IL-10 pathways in fibroblasts and in clinical settings, potentially reducing major adverse cardiovascular events. Further investigation into broader aspects of this pathway, particularly in cardiac fibroblasts, is required.

Vehicular ad hoc networks (VANET) are a subset of mobile ad hoc networks used to communicate between vehicles and infrastructure. During vehicular congestion communication, nodes compete to acquire channels, causing the channels to become congested. The congestion on the vehicle network results in increased delay and packet loss, resulting in reduced VANET performance. To address this problem, we developed a load-aware and priority adaptive traffic congestion control method in vehicular ad hoc networks (VANETs). The proposed scheme identifies less-congested road segments based on the network's load and reduces traffic congestion by suggesting other routes between nearby roadside units (RSUs). This research aims to improve the efficiency of the vehicular environment by utilizing the movement of vehicles with (RSUs) and sharing the traffic load between them. Simulation results demonstrate the effectiveness of the proposed protocol in reducing congestion and enhancing the overall performance of VANETs. To validate the proposed algorithm, we have implemented and tested the proposed algorithm using Network Simulator 3 (NS3) for Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication scenarios and computed the performance of the algorithm on different parameters of the network. The simulation result of the proposed load aware and priority adaptive traffic congestion control method in VANET improved the packet delivery ratio, packet lost ratio, and end-to-end delay by 96%, 4.1%, and 1102 milliseconds, when compared to TDCCA value of 92%, 5.7%, and 1154 milliseconds, respectively

Three-phase induction motors play a key role in industrial operations. However, their failure can result in serious operational problems. This study focuses on the early identification of faults through accurate diagnosis and classification of faults in three-phase induction motors, using artificial intelligence techniques, by analyzing current, temperature, and vibration signals. Experiments were conducted on a test bench, simulating real operating conditions, including stator phase unbalance, bearing damage and shaft unbalance. To classify the faults, an Auto-Regressive Neural Network with Exogenous Inputs (NARX) was developed. The parameters of this network were determined through a process of selecting the best network, using the scanning method with multiple training and validation iterations, with the introduction of new data. The results of these tests showed that the network had excellent generalization in all the situations evaluated, achieving the following accuracy rates: Motor without fault = 94.2%, Unbalance fault = 95%, Bearings with fault =98% and Stator with fault = 95%.

Evolutionary algorithms and swarm intelligence algorithms find applicability in reinforcement learning of neural networks due to their independence from gradient-based methods. To achieve successful training of neural networks using these algorithms, careful considerations must be made to select appropriate algorithms due to the availability of various algorithmic variations. In Part1, the author previously reported experimental evaluations on Evolution Strategy　for reinforcement learning of binary neural networks, utilizing the Pendulum control task. This article constitutes Part2 of the series of comparative research. In this study, Genetic Algorithm is adopted as another evolutionary algorithm. Wilcoxon signed rank test revealed that there was no statistically significant difference between the fitness scores obtained using GA and those using ES. However, the p-value 0.11 indicated that GA worked better than ES on this training task. As the values of binary weights, {-1, 1} were significantly superior to {0, 1} (p < .01). The motion of the pendulum controlled by the binary MLP after the training showed that the binary MLP successfully swung the pendulum swiftly into an inverted position and maintained its stability after inversion.

Currently, foam mixers are predominantly categorized into two types: low-pressure and high-pressure. Low-pressure mixers primarily use agitator rotation to mix materials. However, they face challenges such as difficulty in cleaning and complex structures. High-pressure mixers, widely adopted for their structural simplicity and lack of need for cleaning devices, effectively mitigate issues associated with low-pressure mixers. Nonetheless, they encounter problems with uneven mixing when processing high-viscosity materials. Historically, the primary solution to this issue was increasing the working pressure. However, the capacity to increase pressure is finite, and higher pressure demands higher quality materials for the mixers. To address the issues high-pressure mixers face when dealing with high-viscosity materials, this study focuses on the polyimide high-pressure mixer. After reviewing previous studies, four design variables were selected: the mixer's impingement angle, inlet diameter, outlet diameter, and impingement pressure. Using a Full Factorial Design of Experiments (DOE), the significance of the impact exerted by these four variables on Mixing Unevenness was analyzed, determining the extent of their influence. Optimal Latin Hypercube Sampling generated several random sample points, followed by the establishment of both a Back Propagation (BP) Neural Network approximation model and a Kriging approximation model. A Genetic Algorithm was then applied to optimize these two approximation models, resulting in the optimal combination of design variables. The research findings indicate that the BP Neural Network exhibits higher predictive accuracy compared to the Kriging prediction model. The optimal combination of parameters, optimized through the Genetic Algorithm (GA), reduced the Mixing Unevenness by 69.1%, effectively decreasing the Mixing Unevenness of the mixer head.

The article discusses issues related to the optimization of the operation of medium voltage networks, focusing on the optimization of network division points and the selection of the optimal network configuration aimed at reducing power and energy losses. The article provides a critical review of selected scientific works from recent years on the issue of network configuration optimization. The developed solution uses non-deterministic network models and statistical and probabilistic approaches. Optimization studies were carried out using heuristic optimization methods and proprietary algorithms for forecasting the load and power generated from renewable energy sources. The solution proposed in the article has many application possibilities. It can be used to optimize the location of network division points (reducing economic costs related to the modernization of network infrastructure) as well as to optimize the reconfiguration process in order to reduce power losses while maintaining voltage criteria.

This study explores the potential of using a Siamese Network as a biomarker for assessing the effectiveness of Dolphin-Assisted Therapy (DAT) in children with Cerebral Palsy (CP). The problem statement revolves around the need for objective measures to evaluate the impact of DAT on patients with CP, considering the subjective nature of traditional assessment methods. The methodology involves training a Siamese network, a type of neural network designed to compare similarities between inputs, using data collected from CP patients undergoing DAT sessions. The network is trained to learn patterns and similarities between pre- and post-therapy evaluations, in order to identify biomarkers indicative of therapy effectiveness. Notably, the Siamese Network’s architecture ensures that comparisons are made within the same feature space, allowing for more accurate assessments. The results of the study demonstrate promising findings, indicating different patterns in the output of the Siamese Network that correlate with improvements in symptoms of CP post-DAT. These findings suggest the potential utility of the Siamese Network as a biomarker to monitor therapy outcomes in children with CP who undergo DAT, offering a more objective and quantifiable approach to assessing therapeutic interventions. Significant disparities in neuronal voltage perturbations, with a mean of 7.9825 dB in specific samples compared to 6.2838 dB in the data set, indicate a marked difference from resting activity. This suggests that dolphin-assisted therapy selectively activates specific brain regions uniquely during the intervention.

This review explores the potential of Intelligent Reflecting Surfaces (IRS) to address limitations of traditional antennas and empower next-generation wireless networks. IRS technology offers a paradigm shift by passively manipulating radio waves, enhancing signal coverage, network capacity, and energy efficiency. IRS technology utilizes flat surfaces equipped with tiny, adjustable reflectors. Unlike traditional antennas that actively send and receive radio waves, IRS passively manipulates incoming signals by reflecting them in specific directions with controlled adjustments. This approach significantly improves signal coverage and strength, particularly in areas with poor reception. The review reveals the IRS technology offers significant advantages over traditional antennas. It can enhance signal coverage by reflecting radio waves, boost network capacity which allows networks to handle a larger number of users simultaneously, promote lower energy consumption compare, and unlock potential of 6G Networks and beyond. Additionally, IRS enables functionalities such as enhanced localization and ultra-fast data transfer. Integrating IRS with next-generation antennas unlocks new possibilities. This includes reconfigurable antennas that dynamically adjust radiation patterns, compact and low-profile designs, and multi-functional antennas that support various services simultaneously. In conclusion, this paves the way for a new generation of robust and sustainable wireless communication systems.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a positive sense single stranded RNA genome containing virus which has infected millions of people all over the world particularly China. The virus has been mutating rapidly enough resulting in the emergence of new variants and sub-variants which have reportedly been spread from Wuhan city of China, the epicenter of the virus, to the rest of China and all over the world. The occurrence of mutations in the viral genome especially in the viral spike protein region has resulted in the evolution of multiple variants and sub-variants which gives the virus benefit of host immune evasion and render the modern-day vaccines and therapeutics ineffective. Therefore, there is a continuous need to study the genetic characteristics and evolutionary dynamics of the SARS-CoV-2 variants. Hence, in this study, a total of 833 complete genomes of SARS-CoV-2 variants (including the wild-type or reference sequence) from Taiyuan and Wuhan cities of China were genetically characterized and studies their phylogenetic and evolutionary dynamics using phylogenetics, genetic similarity, and phylogenetic network analysis. Our results show a lot of useful information regarding the evolutionary dynamics of the viral variants, their genetic similarities/dissimilarities, and their phylogenetic relationships. It has been reported in this study that the SARS-CoV-2 variants have nine recombinant events which suggest very important information regarding the viral evolution. In addition, the phylogenetic network analysis shows the number of mutations resulting in emergence and clustering of the viral variants of both cities which reveal significant information regarding the phylodynamics of the virus. This study, to the best of our knowledge, is the first ever genetic comparative study of Taiyuan and Wuhan cities. This study will help better understand the virus, cope with the emergence, and spread of new variants at local level as well as at international level and inform the public health authorities to make better informed decisions in designing new viral vaccines and therapeutics.

The increasing use of Convolutional Neural Networks (CNN) has brought about a significant transformation in numerous fields, such as image categorization and identification. In the development of a CNN model for classifying images of sericea lespedeza (SL; Lespedeza cuneata) from weed images, four architectures were explored: CNN-Model Variant 1, CNN-Model Variant 2, VGG16, and ResNet50. The CNN-Model Variant 1 demonstrated 100 % validation accuracy, while Variant 2 achieved 90.78% validation accuracy. Pre-trained models, like VGG16 and ResNet50, were also analyzed. In contrast, ResNet50's steady learning pattern indicated potential for better generalization. A detailed evaluation of these models revealed that Variant 1 achieved a perfect score in precision, recall, and F1-score, indicating superior optimization and feature utilization. Variant 2 presented a balanced performance, with metrics between 86% and 93%. The VGG16 mirrored the behavior of Variant 2, both maintaining around 90% accuracy, but ResNet50's results revealed a conservative approach for class 0 predictions. Overall, Variant 1 stood out in performance, while both Variant 2 and VGG16 showed balanced results. The reliability of CNN model Variant 1 is highlighted by the significant accuracy percentages, which demonstrate its potential for practical implementation in agriculture. Smartphone application for the identification of SL in a field-based trial has shown promising results with an accuracy of 98%-99%. Using a CNN model with batch normalization has the potential to play a crucial role in redefining and optimizing the management of undesirable vegetation in the future.

Time series forecasting has been a challenging area in the field of Artificial Intel-ligence. Various approaches such as linear Neural Networks, Recurrent Linear Neural Networks, Convolutional Neural Networks and recently Transformers have been attempted for the time series forecasting domain. Although, Trans-former-based architectures have been outstanding in the Natural Language Pro-cessing domain, especially in autoregressive language modeling, the initial at-tempts to use transformers in the time series arena have met with mixed success. One of the recent papers demonstrated that modeling the multi-variate time se-ries problem as independent channels on a transformer architecture produced lower mean square and mean absolute errors on benchmarks, however, another recent paper claimed that a simple linear neural network out-performs previous approaches including transformer-based designs. We investigate this paradox in detail comparing the linear Neural Network and Transformer based designs, provide insights to why a certain approach may be better for a particular type of problem. We also improve upon the recently proposed simple linear Neural Network based architecture by using dual pipelines with batch normalization and reversible instance normalization. Our enhanced architecture outperforms all existing architectures for time series forecasting on majority of the popular benchmarks.

Camouflage evaluation typically involves human visual search and detection experiments that are time-consuming and expensive. Hence, there is a need for models that compute camouflage effectiveness from digital imagery. Convolutional neural networks are a powerful tool for automatic object detection and recognition. We investigated whether such a network (YOLO) can also provide a measure of camouflage effectiveness that is related to human perception. To this end, human performance measures of camouflage effectiveness such as detection time and target conspicuity were obtained in observer experiments and compared with the performance of the (standard, i.e. pre-trained) YOLO-V4-tiny algorithm. YOLO provides the probability that a detected object is correctly classified, and this is adopted as our measure of camouflage effectiveness. We compared the YOLO-predicted classification probability for a person in camouflage clothing moving through rural and urban scenes to human detection performance. Overall, camouflage effectiveness predicted by YOLO correlates with human observer performance. At close distances, YOLO’s classification performance appears sensitive to high-contrast shape-breaking elements of a camouflage pattern. This suggests that YOLO may be adapted to assess camouflage effectiveness for a wide range of applications, such as evaluating or optimizing one’s signature and predicting optimal hiding locations in a given environment. Further research is required to fully establish YOLO’s limitations and applicability for this purpose.

Reasonable allocation of production plans and saving the overall electricity cost are crucial for large manufacturing conglomerates, which is an effective way for conglomerates to realize open-source and cost-saving. This paper develops an optimization model of off-site industrial production scheduling for enterprises in response to the problems of high electricity costs due to the irrational allocation of production schedules on the demand side of China's electric power and the difficulty of promoting industrial and commercial distributed photovoltaic (PV) projects in China. The model makes full use of the conditions of different PV resources and different electricity prices in different places to optimize the scheduling of industrial production in different places. The model is embedded with two sub-models of the electricity price prediction model and distributed photovoltaic kWh cost model to complete the model parameters, in which the electricity price prediction model utilizes the LSTM neural network. The integrated model is then solved by the particle swarm algorithm. To verify the effectiveness of the model in solving the problem of off-site production scheduling on the demand side of electricity, the research team researched two off-site pharmaceutical factories belonging to the same large pharmaceutical company. Finally, the research data were substituted into the model for solving, and it was concluded that the optimization model could significantly reduce the cost of electricity consumption of the enterprise by about 7.9 %.

Military medical service is an important component of the Armed Forces of any country [1]. In wartime, healthcare performs critical functions of medical support for military operations, providing effective and qualified care to the wounded and sick. Consequently, it is imperative to equip military hospitals with state-of-the-art materials and technical resources commensurate with the predefined requirements. In this study, we examine a series of enhancements applied to the TurtleBot 3 Waffle Pi-based medical assistant robot, tailored for deployment in military hospital settings.

Blockchain technology has emerged as a necessity for the decentralization of payment methods and transactions, but it has brought with it many properties of distributed systems that have made it a crucial technology for overcoming some of society's challenges, especially in the context of decentralizing services, transparency of information, availability, and security. Its architecture and communication methods, although possessing some complex nuances to understand, particularly for the lay audience in the field of distributed systems, protocols, and computer networks. In this article, we will explore some topics of distributed systems related to blockchain technology.

Taekwondo has evolved from a traditional martial art into an official Olympic sport. This study introduces a novel action recognition model tailored for Taekwondo unit actions, utilizing joint motion data acquired via wearable inertial measurement unit (IMU) sensors. The utilization of IMU sensor-measured motion data facilitates the capture of the intricate and rapid movements characteristic of Taekwondo techniques. The model, underpinned by a conventional convolutional neural network (CNN)-based image classification framework, synthesizes action images to represent individual Taekwondo unit actions. These action images are generated by mapping joint motion profiles onto the RGB color space, thus encapsulating the motion dynamics of a single unit action within a solitary image. To further refine the representation of rapid movements within these images, a time-warping technique was applied, adjusting motion profiles in relation to the velocity of the action. The effectiveness of the proposed model was assessed using a dataset compiled from 40 Taekwondo experts, yielding remarkable outcomes: an accuracy of 0.998, precision of 0.983, recall of 0.982, and an F1-score of 0.982. These results underscore the time-warping technique's contribution to enhancing feature representation, as well as the proposed method's scalability and effectiveness in recognizing Taekwondo unit actions.

Wireless Sensor Networks (WSNs) are gaining traction in the realm of network communication, renowned for their adaptability, configuration, and flexibility. The forthcoming network traffic within WSNs can be forecasted through temporal sequence models. In this correspondence, We present a method (TSENet) that can accurately predict the traffic in the cellular network. TSENet is composed of transformers and self-attention network. We have designed a temporal transformer module specifically for extracting temporal features. This module accomplishes this by modeling the traffic flow within each grid of the communication network at both near-term and periodical intervals. Simultaneously, we amalgamate the spacial features of each grid with information from its correlated grids, generating spacial predictions within the spacial transformer. Furthermore, we employ self-attention aggregation to capture dependencies between external factor features and celluar data features. Empirical assessments performed on a genuine cellular traffic dataset offer compelling evidence substantiating the efficacy of TSENet.

Evolutionary algorithms find applicability in reinforcement learning of neural networks due to their independence from gradient-based methods. To achieve successful training of neural networks using evolutionary algorithms, careful considerations must be made to select appropriate algorithms due to the availability of various algorithmic variations. In Part1 and Part2, the author previously reported experimental evaluations on Genetic Algorithm and Evolution Strategy for reinforcement learning of neural networks, utilizing the Acrobot control task. This article constitutes Part3 of the series of comparative research. In this study, Differential Evolution is adopted as the third instance of major evolutionary algorithms. The experimental results show a statistically significant superiority of DE over both GA and ES (p < .01). In addition, DE exhibits its robustness to variations in hyperparameter configurations (the number of offsprings and generations). In the previous experiments, both ES and GA showed significant performance differences depending on the configurations, whereas in the experiment reported in this article, such differences are not detected for DE.

This paper reviews several techniques, including analytical theories and real frequency techniques, to design ultra-wideband impedance matching networks for low-power applications. The comparison of solutions obtained using the simplified real frequency technique to the published solution using Chebyshev filter theory is demonstrated.

This study explores the use of Bayesian Neural Networks (BNNs) for estimating chlorophyll-a concentration ([CHL-a]) from remotely sensed data. The BNN model enables uncertainty quantification, offering additional layers of information compared to traditional ocean colour models. An extensive in situ bio-optical dataset is utilized, generated by merging 27 data sources across the world’s oceans. The BNN model demonstrates remarkable capability in capturing mesoscale features and ocean circulation patterns, providing comprehensive insights into spatial and temporal variations in [CHL-a] across diverse marine ecosystems. In comparison to established ocean colour algorithms, such as Ocean Colour 4 (OC4), the BNN shows comparable performance in terms of correlation coefficients, errors, and biases when compared with the in situ data. The BNN, however, further provides critical information about the distribution of [CHL-a], which can be used to assess uncertainties in the prediction. Moreover, the BNN model’s ability to provide more information for coastal waters, especially with higher spatial resolution imagery, offers valuable advantages for marine research and management. By quantifying uncertainty, the model builds confidence in [CHL-a] predictions, even in regions with limited data coverage. This is reflected by the spread in the model predictions, where the BNN can detect a range of uncertainty bounds that reflect confidence in the retrievals. Introducing BNNs as probabilistic ocean colour models represents a significant advancement, enabling more accurate and reliable predictions of [CHL-a] in the global oceans.

Neural-objective and image optimization approaches for drilling fluid rheology automation are crucial for drilling engineering optimization. A myriad of intelligent computational models are employed to predict and monitor the parameters of mud rheology and filter cake permeability posture using an artificial neural network feedforward (ANN-FF) function, a non-ANN-FF function, an image processing tool, and a model optimization tool. 498 datasets of synthetic-based mud (SBM) flat rheology from a drilling mud laboratory at Nazarbayev University were imported into a user-friendly MATLAB application using image processing and nftool. Apart from the Google TensorFlow Sequential API DNN architecture used, a Levenberg-Marquardt training algorithm with input sigmoid hidden neurons 10, 12, and 18 coupled with a linear output layer was also used to predict the SBM flow index. OBM and SBM filter cakes were processed for void spaces; however, a model study was extrapolated to maximize filtration volume Vf. The study’s findings show that the ANN-FF model employed for rheological property monitoring and prediction had higher steady and exponential levels of accuracy and a correlation coefficient of 0.96-0.99. More so, SBM and OBM image processing presented an area of void spaces of 1790M2 and 1739M2, respectively. The porosity and permeability postures of both SBM and OBM resulted in a significant void space capable of maximizing the flow index. In addition, the single-objective modelling based on a genetic algorithm did validate the experimental rheological data for flow index or filtration volume Vf optimization; the study resulted in the finding that the constrained physics-informed objective function hindered maximizing oil recovery and instead predicted possible formation damage. It is empirical to note that automating flow predictions with neural, objective, and image functions has presented an alternative novel method for non-programmers using MATLAB and Google Colabs that is capable of enhancing mud rheological parameters, drilling efficiency, and hydrocarbon recovery.

In our digital era, digital platforms become an unprecedented opportunities for manufacturing enterprises to leverage their business strategy. This study investigates how manufacturing enterprises can enhance organizational duality through digital platforms. Specifically, this study examines the effect of digital platforms and capability reconfiguration on entrepreneurial organizational duality. The study also examines how network digital atmosphere moderates this relationship. Based on analysis of 286 manufacturing enterprises undergoing digital transformation, the results indicate that digital platforms positively affect both organizational duality and the capability reconfiguration. The results indicate that digital platforms have a positive indirect effect on organizational duality via evolutionary capability reconfiguration and substitutional capability reconfiguration, respectively. The study also shows that network digital atmosphere moderate this effect.These findings contribute theoretically to the research on digital platforms and manufacturing enterprises while also providing important managerial implications for policymakers and manufacturing enterprises to help firms implement digital strategies.

In the pursuit of energy-efficient spiking neural network (SNN) hardware, synaptic devices lev-eraging emerging memory technologies hold significant promise. This study investigates the ap-plication of the recently proposed HfO2/SiO2-based interface dipole modulation (IDM) memory for synaptic spike timing-dependent plasticity (STDP) learning. Firstly, through pulse measure-ments of IDM metal-oxide-semiconductor (MOS) capacitors, we demonstrate that IDM exhibits an inherently nonlinear and near-symmetric response. Secondly, we discuss the drain current re-sponse of a field-effect transistor (FET) incorporating a multi-stack IDM structure, revealing its nonlinear and asymmetric pulse response, and suggest that the degree of the asymmetry depends on the modulation current ratio. Thirdly, to emulate synaptic STDP behavior, we implement dou-ble-pulse-controlled drain current modulation of IDMFET using a simple bipolar rectangular pulse. Additionally, we propose a double-pulse-controlled synaptic depression that is valuable for optimizing STDP-based unsupervised learning. Integrating the pulse response characteristics of IDMFETs into a two-layer SNN system for synaptic weight updates, we assess training and classification performance on handwritten digits. Our results demonstrate that ID-MFET-based synaptic devices can achieve classification accuracy comparable to previously re-ported simulation-based results.

Survival models have become popular for credit risk estimation. Most current credit risk survival models use an underlying linear model. This is beneficial in terms of interpretability but is restrictive for real-life applications since it cannot discover hidden nonlinearities and interactions within the data. This study uses discrete time survival models with embedded neural networks as estimators of time to default. This provides flexibility to express nonlinearities and interactions between variables, and hence allows for models with better overall model fit. Additionally, the neural networks are used to estimate Age-Period-Cohort (APC) models so that default risk can be decomposed into time components for loan age (maturity), origination (vintage) and environment (e.g., economic, operational and social effects). These can be built as general models, or as local APC models for specific customer segments. The local APC models reveal special conditions for different customer groups. The corresponding APC identification problem is solved by a combination of regularization and fitting the decomposed environment time risk component to macroeconomic data, since the environmental risk is expected to have a strong relationship with macroeconomic conditions. Our approach is shown to be effective when tested on a large publicly available US mortgage data set. This novel framework can be adapted by practitioners in the financial industry to improve modelling, estimation and assessment of credit risk.

River bed materials serve multiple environmental functions as a habitat for aquatic invertebrates and fishes. At the same time, the particle size of the bed material reflects the tractive force of the flow regime in a flood and provides useful information for flood control. The traditional river bed particle size surveys, such as sieving, require time and labor to investigate river bed materials. The authors proposed a method to classify aerial images taken by unmanned aerial vehicle (UAV) using convolutional neural networks (CNN), our previous study showed that terrestrial riverbed material could be classified with high accuracy. In this study, we attempted to classify riverbed materials distributed in shallow waters where the bottom can be seen from UAVs. After training the CNN to classify the images with the same grain size as being in the same class even if the surface flow types taken overlapping the riverbed material were different, the total accuracy reached 90.3%. Moreover, the proposed method was applied to the wide-ranging area to determine the distribution of the particle size. In parallel, the microtopography was surveyed using Lidar-UAV, and the relationship between the microtopography and particle size distribution was discussed. In the steep section, coarse particles were distributed and formed a rapid. Fine particles were deposited on the upstream side of those rapids, where the slope had become gentler due to the damming. There was good agreement between the microtopographical trends and the grain size distribution.

Wearing gloves while operating machinery in workshops is an essential precaution to prevent mechanical injuries and burns from high temperatures, among other potential hazards. Ensuring workers are properly equipped with gloves, which is a crucial measure in accident prevention. Glove images often occupy a minimal proportion of the frame and are easily obscured by cluttered backgrounds, especially with limited edge computing resources. Consequently, this study proposes a glove detection algorithm called YOLOv8-AFPN-M-C2f based on YOLOv8, offering swifter detection speeds, lower computational demands, and enhanced accuracy for workshop scenarios. This research innovates by substituting the head of YOLOv8 with the AFPN-M-C2f network, amplifying the pathways for feature vector propagation, and mitigating semantic discrepancies between non-adjacent feature layers. Additionally, the introduction of a superficial feature layer enriches surface feature information, augmenting the model's sensitivity to smaller objects. To assess the performance of the YOLOv8-AFPN-M-C2f model, we conducted multiple experiments using a factory glove detection dataset compiled for this study. The results indicate that the enhanced YOLOv8 model surpasses other network models. Compared to the baseline YOLOv8 model, the refined version shows a 2.6% increase in mAP@50%, a 90.1% rise in FPS, and a 13% reduction in the number of parameters. This research contributes an effective solution for the detection of glove adherence.

Evolutionary algorithms find applicability in reinforcement learning of neural networks due to their independence from gradient-based methods. To achieve successful training of neural networks using evolutionary algorithms, careful considerations must be made to select appropriate algorithms due to the availability of various algorithmic variations. The author previously reported experimental evaluations on Evolution Strategy for reinforcement learning of neural networks, utilizing the Acrobot control task. In this study, Genetic Algorithm is adopted as another instance of major evolutionary algorithms. Experimental results demonstrate that there was no statistically significant difference between the experimental performances of GA and ES, but the priority of generations and offsprings was different; GA performed better with a greater number of generations while ES performed better with a greater number of offsprings. Eight hidden units were the best among four variations (4, 8, 16 or 32 units), which aligns with previous study using ES.

Analyzing network robustness against random failures or malicious attacks is a critical research issue in network science as it helps to enhance the robustness of beneficial networks or efficiently disintegrate harmful networks. Most studies commonly neglect the impact of the attack success rate (ASR) and assume that attacks on the network will always be successful. However, in real-world scenarios, an attack may not always succeed. This paper proposes a novel robustness measure called Robustness-ASR (RASR), which utilizes mathematical expectations to assess network robustness when considering the ASR of each node. To efficiently compute the RASR for large-scale networks, a parallel algorithm named PRQMC is presented, which leverages randomized quasi-Monte Carlo integration to approximate the RASR with a faster convergence rate. Additionally, a new attack strategy named HBnnsAGP is introduced to better assess the lower bound of network RASR. Finally, the experimental results on 6 representative real-world complex networks demonstrate the effectiveness of the proposed methods compared with the state-of-the-art baselines.

A long time has passed since COVID-19 was discovered and widely disseminated. Both machine learning and deep learning have moved towards the research of diagnosing COVID-19. Compared to deep learning, traditional machine learning-based methods can also achieve good diagnostic results if they are improved based on innovative points. And deep learning remains a more popular research object. Based on the deep neural network in the field of deep learning, the convolutional neural network, recurrent neural network, long and short-term memory network, and Transformer model have been extended. These models improve the performance and processing power of neural networks by introducing new structures and algorithms. This paper will introduce the basic concepts of machine learning and deep learning, as well as the details of using the relevant methods of both to help diagnose COVID-19.

In recent years web applications and on-line business transactions have grown many folds. Consequently, also cyberattacks have increased and represent a serious threat to the pervasive digital services upon which our society relies. To mitigate cyberattacks, many countermeasures are deployed on computing nodes (e.g., anti-malware software) as well as on network devices to detect and possibly block malicious packets in transit; these monitoring devices broadly go under the name of firewalls. Firewalls are designed according to two main architectural approaches: software running on a standard or embedded computer, or purposedly designed hardware, e.g., ASICs. Software-based solutions have the advantage of high flexibility and can be ported on easily upgradable hardware. However, hardware implementation represents the only viable solution for high data rates. On the market, very fast devices of the latter kind are available, but their cost is typically very high, especially considering that their ultra-optimized design makes updating them very difficult, with the consequence of a rather short lifespan. As a more balanced alternative, we wanted to investigate the use of an FPGA architecture, , which is significantly easier to update than custom-built chips, and features low-latency and high-throughput characteristics concurrently, making it preferrable to other programmable systems based on GPUs or microcontrollers. In this paper a packet sniffer that has been designed on FPGA with a 1 Gbit/s data transfer rate is presented. The system is implemented on the FPGA development board KC705 by Xilinx, can analyze Ethernet frames, checking the frame fields against a set of rules defined by the user and calculates statistics of the received Ethernet frames over time. The designed packet sniffer has been successfully tested both with Ethernet frames ad hoc generated using a packets generator, and with real web traffic by connecting the packet sniffer to the internet.

Primary central nervous system and brain tumors are one of the global burdens that are continuously increasing in cases and requiring more treatment options. Surgery has been the leading treatment option for tumors, however, the localization of the tumor and its infiltrating nature make this option challenging. DNA microarray expression profiles of different CNS tumors provide insight into potential biomarkers for identifying different tumor types and subtypes. Here, we utilized the differentially expressed genes common in four expression profiles, GSE66354, GSE68848, GSE74195, and GSE43290 to reconstruct the gene co-expression network. In this study, we were able to identify preserved cluster genes, hub genes, co-regulating transcription factors, miRNA families, and candidate repurposed drugs. Fourteen identified hub genes, which were, CACNA1A, DNM1, GABRA1, GRIA2, MAPT, SLC17A7, SNAP25, SNAP91, STXBP1, SYT1, COL1A1, COL6A2, FBN2, and FN1 appeared to play a role in tumor progression and may serve as drug targets. We also reported the DEGs in each tumor type, which were ATRT, EPN, PA, MED, PNET, MEN, ACM, ODG, and GBM. Five identified miRNA families, which were, let-7 family, mir-124 family, mir-1 family, mir-103 family, and mir-27 family described in the literature to have tumor-suppressing characteristics. Drug-gene and drug-transcription factor network revealed 32 candidate repurposed drugs and 13 were validated through connectivity map analysis. Here, two main repurposed drugs fit the regulatory requirement for gene interaction, these were, quercetin and vorinostat. Future validation in experimental studies may utilize the future use of the candidate repurposed drugs. Our study provides insights into drug repurposing prospects and understanding tumor expressions.

With the deepening of the diversification and openness of financial system, financial vulnerability, as an endogenous attribute of financial system, becomes an important measurement of financial security. Based on network analysis, we introduce network curvature indicator improved by Copula entropy as an innovative metric of financial vulnerability. Compared with previous network curvature analysis method, CE-based curvature proposed in this paper can measure market vulnerability and systematic risk with significant advantages.

Big data technologies are disruptive technologies that affect every business, including those in the construction industry. The Thai government has also been affected, and attempted to use machine learning techniques with the analytics of big data technologies to predict which construction projects have a winning price over the project budget. However, this technology was never developed, and the government did not implement it because they had data obtained via a traditional data collection process. In this study, traditional data were processed to predict behavior in Thai government construction projects using a machine learning model. The data were collected from the government procurement system in 2019. There were seven input data, including project owner department, type of construction project, bidding method, project duration, project level, winning price over estimated price, and winning price over budget. A range of classification techniques, including an artificial neural network (ANN), a decision tree (DC), and K-nearest neighbor (KNN), were used in this study (ANN). According to the results, after hyperparameter tuning, ANN had the greatest prediction accuracy with 78.9 percent. This study confirms that data from the Thai government procurement system can be investigated using machine learning techniques from big data technologies.

Evolutionary algorithms find applicability in the reinforcement learning of neural networks due to their independence from gradient-based methods. To achieve successful training of neural networks using evolutionary algorithms, careful considerations must be made to select appropriate algorithms due to the availability of various algorithmic variations. The author previously reported experimental evaluations on Evolution Strategy for reinforcement learning of neural networks, utilizing the pendulum control task. In this study, the Acrobot control task is adopted as another task. Experimental results demonstrate that ES successfully trained a Multi-Layer Perceptron to achieve a remarkable height of 99.85% concerning the maximum height. However, the trained MLP failed to maintain the chain end in an upright position throughout an episode. In this study, it was observed that employing 8 hidden units in the neural network yielded better results with statistical significance compared to using 4, 16, or 32 hidden units. Furthermore, the findings indicate that a larger population size in ES led to a more extensive exploration of potential solutions over a greater number of generations, which aligns with the previous study.

Near Earth Asteroids represent potential threats to human life because their trajectories may bring them in the proximity of the Earth. Monitoring these objects could help predict future impact events, but such efforts are hindered by the large numbers of objects that pass through the Earth’s vicinity. Additionally, there is also the problem of distinguishing asteroids from other objects in the night sky, which implies sifting through large sets of telescope image data. Within this context, we believe that employing machine learning techniques could greatly improve the detection process by sorting out the most likely asteroid candidates to be reviewed by human experts. At the moment, the use of machine learning techniques is still limited in the field of astronomy and the main goal of the present paper is to study the effectiveness of deep CNNs for the classification of astronomical objects, asteroids in this particular case, by comparing some of the well-known deep convolutional neural networks, including InceptionV3, Xception, InceptionResNetV2 and ResNet152V2. We have applied transfer learning and fine-tuning on these pre-existing deep convolutional networks and from the results that we have obtained one can see the potential of using deep convolutional neural networks in the process of asteroid classification. The InceptionV3 model has the best results in the asteroid class, meaning that by using it, we loose the least number of valid asteroids.

Stemming from the object overlap and undertraining from the few samples, the road dense object detection is confronted with the poor object identification performance and the inability to recognize edge objects. Based on this, one transfer learning-based YOLOv3 approach for identifying dense objects in the road has been proposed. Firstly, Darknet-53 network structure is adopted to obtain pre-trained YOLOv3 model, then the transfer training is introduced as the output layer for the special dataset of 2000 images containing vehicles; in the proposed model, one random function is adapted to intialize and optimize the weights of the transfer training model, which is seperately designed from the pre-trained YOLOv3; and the object detection classifier replaces the fully connected layer, which further improves the detection effect, the reduced size of the network model can further reduce the training and detection time, and can be better applied to actual scenarios. The experimental results demonstrate that the object detection accuracy of the presented approach is 87.75% for the Pascal VOC 2007 dataset, which is superior to the traditional YOLOv3 and the traditional YOLOv2 by 3.05% and 11.15%, respectively. Besides, the test was carried out using UA-DETRAC, a public road vehicle detection dataset, the object detection accuracy of the presented approach reaches 79.23% in detecting images, which is 4.13% better than the traditional YOLOv3, and compared with the existing relatively new object detection algorithm YOLOv5, the detection accuracy is 1.36% better. Moreover, the detection speed of the proposed YOLOv3 method reaches 31.2 Fps/s in detecting images, which is 7.6 Fps/s faster than the traditional YOLOv3, and compared with the existing relatively new object detection algorithm YOLOv5, the speed is 4.3 Fps/s faster; the proposed YOLOv3 performs 79.38Bn of floating point operations per second in detecting video, which obviously surpasses the traditional YOLOv3 and the newer object detection algorithm YOLOv5.

The increase and diversity of low-cost Air Quality (AQ) sensors, as well as their flexibility and low power consumption, offers us the opportunity to integrate them into a broad AQ wireless sensor networks with the aim of enabling real-time monitoring and higher spatial sampling density of pollution in all parts of the cities. Considering that the vast majority of the population lives in these cities and the increase in respiratory/allergic problems in a large part of the population, it is of great interest to offer services and applications to improve their quality of life. In the ECO4RUPA project we focus on this kind of service, proposing an inclusive and intelligent routing ecosystem carried out by using a network of low-cost AQ sensors with the support of 5G communications along with official AQ monitoring stations, assisted with artificial intelligence to improve the AQ monitoring data and by spatial interpolation techniques to enhance its spatial resolution. The goal of this service is to calculate healthy walking and/or cycling routes according to the particular citizen’s profile and needs. We provide and analyse the results of the proposed route planner under different scenarios (different time tables, congestion road traffic and routes) and different user’s profiles, with special interest on citizens with asthma and pregnant women, since both require special needs. In summary, our approach can lead to an approximately average reduction in pollution exposure of 17.82%, while experiencing an approximately average increase in distance travelled of 9.8 %.

Social network analysis is extensively employed for the examination of relationships and networks among actors across diverse disciplines and thematic research domains. Nevertheless, the utilization of this approach in agriculture and natural resources, particularly pertaining to watershed management, has been relatively limited. The objective of this study is to conduct an analysis of the social interaction and networks of Cidanau Watershed Management (CWM) stakeholders and measure the nodes of units as individuals or organizations in relation to the role and influence of each other through a quantitative approach. The results of the data analysis show that Fkdc has a high degree of centrality and centrality betweenness. This shows that Fkdc has a strong influence and controls CWM network interaction. Conversely, Hs (households) have a low degree of centrality, and centrality betweenness values have a low influence and role in actors’ networks. According to the closeness centrality parameter, it is observed that both Households (Hs) and Universities (Univ) have exhibited the shortest geodesic distance in relation to other stakeholders within the network, thus indicating their highest level of closeness. It means that households and universities had interdependency and were controlled by others. The lowest centrality of Fkdc means that Fkdc has independence of influence and high control access over others. There are four CWM network clusters based on total link strength. Fkdc demonstrates outstanding leadership qualities, enjoys a strong reputation, and possesses a high degree of popularity within the network. Furthermore, Fkdc serves as a positive central node that effectively brings together the interests of all actors, fostering a collective commitment towards shared goals and mutual rewards. In the CWM case, Fkdc is a sustainable watershed organization that acts as a central network body and plays a fundamental role as an intermediary between networks and interactions. Ultimately, with regards to the viewpoint of CWM, the decentralization of watersheds aimed to facilitate wider public engagement by establishing Fkdc as a central network entity and assuming a crucial role as a medium for the dissemination of information, exchange of ideas, and the exertion of influence among its members.

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.

This study probes the probabilistic features of major fire hazards in enclosed spaces to establish their importance to the occurrence of fires onboard ro-ro passenger ships and in turn to raise effective operational countermeasures. Distinct from the previous studies, the present research employs Bayesian Network (BN) analysis to determine the probabilities of fire hazards more effectively. The findings of the research include five critical basic events (BE) identified namely, vehicle engine fire (fuel system fault), vehicle electric fire (electrical equipment defect or short circuit), used car electrical fire, reefer units electrical fire (electrical appliances defects or short circuit), and cargo spontaneous combustion. Additionally, the risk of fire for lithium - ion battery powered vehicles is also highlighted in the process of BN analysis, which prompts the authors to propose preventive measures for mitigating the possibility of fire occurrence on this type of electric vehicles. It is hoped that these measures can be essential justifications for establishing relevant rules regarding carrying LIB vehicles in enclosed spaces on international level.

For centuries, Foeniculi fructus (F. fructus) has been used as a traditional herbal medicine in China and Europe and is widely used as a natural therapy for digestive disorders, including indigestion, flatulence, and bloating. The mechanism by which F. fructus alleviates functional dyspepsia was analyzed through network pharmacology, and its therapeutic effect on an animal model of functional dyspepsia was investigated. The compounds, targets, and related diseases of F. fructus were studied utilizing the traditional Chinese medicine systems pharmacology (TCMSP) database. Information on the target genes was classified using the UniProt database. A network was built using Cytoscape 3.9.1, and functional dyspepsia-related genes were checked using the Cytoscape string application. The efficacy of F. fructus on functional dyspepsia, including gastroparesis and gastrointestinal motility, was confirmed by treatment with its extract in a mouse model of loperamide-induced functional dyspepsia. Seven compounds targeted twelve functional dyspepsia-associated genes. In a functional dyspepsia mouse model, F. fructus significantly suppressed the symptoms when compared to that in the control group. Our animal studies showed that the mechanism of action of F. fructus is closely related to gastrointestinal motility. Based on animal experimental results, the results showed that F. fructus provides a potential means to treat functional dyspepsia, suggesting that its medical mechanism for functional dyspepsia can be described by the relationship between seven key compounds of F. fructus, including oleic acid and β-sitosterol, and 12 functional dyspepsia-related genes.

The intrinsic non-linearity and complexity of suspended sediment dynamics, which are impacted by the geographical variability of basin parameters and temporal climatic patterns, make it difficult to estimate suspended sediment concentration (SSC) accurately in hydrological processes. Deep neural networks (DNNs), a cutting-edge modeling method that can capture the innate non-linearity in hydrological systems, have emerged as a solution to this problem. Using primary data on discharge, SSC, and turbidity, the long short-term memory method of DNNs was employed in this study to simulate the discharge-suspended sediment connection for the Esopus Creek, NY, USA as the first objective. To develop effective modeling strategies, multiple scenarios of feature selection, including combinations of discharge, turbidity and SSC for preceding days, were considered. Secondly, the variations in SSC and discharge between the three USGS gages along Esopus Creek that were upper stream, lower stream, and downstream, adjacent to the pourpoint were analyzed. Statistical metrics and scenarios of feature importance were used to evaluate the effectiveness of the DNN-based models. The study shows that among 24 DNN approaches feature selection is crucial in simulating the daily SSC, and discharge and presenting novel research directions for utilizing machine learning algorithms in water quality.

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-driven models based on artificial intelligence are efficiently used to solve complex problems. The quality of groundwater is of utmost importance, as it directly impacts human health and the environment. In major parts of the world groundwater is the main source of drinking water, it is essential to periodically monitor water quality. Conventional water quality monitoring techniques involve periodical collection of water samples and analysis in the laboratory. This process is expensive, time consuming and involves lot of manual labor. The aim of our study is to build an ant colony optimized neural network for predicting groundwater quality parameters. We have proposed artificial neural network comprising of six hidden layers. The approach was validated using our groundwater quality dataset of a hard rock region in the northern part of Karnataka, India. Groundwater samples were collected by us periodically from March 2014 to October 2020 from 50 wells in this region. These samples where analysed for measuring the pH, Electrical Conductivity, Na+, Ca+, Na+, K+, Mg2+, F-, Cl- and U+. The temporal dataset was split for training, testing and validation of our model. Metrics such as R2 (Coefficient of Determination), RMSE (Root Mean Squared Error), NSE (Nash–Sutcliffe efficiencies) and MAE (Mean Absolute Error) were used to evaluate the prediction error and model performance. These performance evaluation metrics indicated the efficiency of our model in predicting the temporal variation in groundwater quality parameters. The method proposed by us can be used for prediction and the temporal frequency of sample collection can be reduced to save time and cost. The results also confirm that the combination of artificial neural network with ACO is a promising tool to optimize weights while training the network, and hence will help in reasonable prediction of groundwater quality parameters.

The rampant misuse of drones poses a serious threat to national security and human life. Currently, the CNN method has been widely used in drone detection. However, there is a challenge that traditional CNN cannot cope with, which is small drone targets often have reduced amplitude or even lost features in infrared images. This paper proposes a Progressive Feature Fusion Network (PFFNet) that gradually increases the response amplitude of the target in the deep network. The Feature Selection Model (FSM) is designed to improve the utilisation of the output coding graph and enhance the feature representation of the target in the network. A lightweight segmentation head is also designed to achieve progressive feature fusion with multi-layer outputs. Experimental results show that the proposed algorithm can achieve low duration and high accuracy in drone target detection. On the public dataset, the IoU is improved by 2.53% and the detection time is reduced by 81.03%.

Cancer is a major concern for people, and accurately predicting the probability of cancer incidence and mortality is an important research topic. With the development of big data and artificial intelligence technology, a new machine learning model has emerged. Using 72,591 pieces of data, including age, case count, population size, race, gender, site of onset, and year of discovery, we built a machine learning model. Through calculations, we found that the decision tree, random forest, logistic regression, support vector machine, and neural network achieved testing accuracies of 62.11%, 61.68%, 54.53%, 55.72%, and 63.10%, respectively. With the help of this model, scientists and policymakers can accurately predict future cancer incidence and mortality rates through databases, allowing them to make relevant policies (such as timely allocating doctors and medical resources) to better serve the people.

Limited accessibility in some parts of Indonesia causes difficulties in logistic distribution, especially for emergency and medical supplies. Emerging drone cargo technology is a potential solution to improve logistic distribution in those areas. However, implementing drone cargo technology involves unknown risks, both in technical and non-technical aspects. Since data on drone operations in Indonesia is limited, a new method is explored to build a Bayesian Network (BN) model for risk analysis of drone crashes in Indonesia’s outermost and underdeveloped areas. The method optimizes the modelling process, in which significant risk factors are selected based on three drone operator companies’ experiences, which include wind speed, rain intensity, and system component failure or malfunction. Real wind speed and rain probability data are then implemented in the model. The operator’s data shows that wind speed contributes to drone crashes, which can be appropriately modelled in the BN model. The model produced a probability of safe operation of 94.1%, comparable to the annual operator’s data. The result shows that most operations are safe, with a minimum case of crashing and no case of harming human life.

Diabetes mellitus is a popular life-threatening disease and patients may gradually have started suffering from other diabetes-causing diseases such as heart attacks, stroke, hypertension, blurry vision, blindness, foot ulcer, amputation, kidney damage and other organ failures before diagnosis. Early detection can help reduce the fatality of this disease. Deep learning models have proven very useful in disease detection and computer-aided diagnosis. In this work, we proposed a deep unsupervised machine learning model for early detection of diabetes using voting ensemble feature selection and deep belief neural networks (DBN). Dataset was obtained from an online repository containing responses of prediagnosed patients to direct questionnaires administered in Sylhet Diabetes Hospital in Sylhet, Bangladesh. The dataset was preprocessed and preprocessed. Features were reduced using the ensemble feature selector. The DBN model was pretrained and tuned to obtain optimal performance. The model was also compared with other models with no multiple hidden layers. The DBN performed at its relative best with F1-measure, precision and recall of 1.00, 0.92 and 1.00 respectively. We conclude that DBN is a useful tool for an unsupervised early prediction of Type II diabetes mellitus.

Herdsmen use different techniques, as per varying geographies and cultures, to keep the cohesion within herds and avoid animals getting lost or predated. However, there is no study on the social behaviour of yaks and on herdsmen management practices. Therefore, this ethology study was initiated by ethnographic inquiries. In Manang, the success of the shepherd is dictated by his personal attribute of “Khula man” or open-heartedness. This attribute refers to good intentions and emotions such as empathy that allows the shepherd to focus more on others than on ownself. This cultural method of assessing the skills required to become a successful and knowledgable shepherd guided us to study the effect of cultural values on the herd’s social behavior. We collected data from two herds living at the same settlement (Yak kharka, 4,100 m altitude, Nepal) by equipping them with loggers. One of the herdsman used the tether rope while other did not. Moreover, the Thaku herd had a more proactive shepherd than the Phurba one. In each herd, 17 animals were equipped with one actigraph wgt3x-BT to measure activity using accelerometer and spatial associations using proximity recorder. One of the herds was equipped with GPS (N=11) as well. Using GPS locations and activity, we showed that the two herds were cohesive and synchronised their activities but the herd with the tether rope was more cohesive. The shepherds also have personal knowledge of the social relationships of their herds and use these relationships to keep the group cohesive and to well manage cattle.

We propose a forward-only multi-layer encoding-decoding framework based on the principle of Maximal Coding Rate Reduction (MCR$^2$), an information-theoretic metric that measures a statistical distance between two sets of feature vectors up to the second moment. The encoder directly transforms data vectors themselves via gradient ascent to maximize the MCR$^2$ distance between different classes in the feature space, resulting in class-wise mutually orthogonal subspace representations. The decoder follows a process symmetric to the encoder, and transforms the subspace feature vectors via gradient descent to minimize the MCR$^2$ distance between the reconstructed data and the original data. We show that the encoder transforms data to linear discriminative representations without breaking the higher-order manifolds, and the decoder reconstructs the data with high fidelity.

Biochemical oxygen demand (BOD) is one of the most important factors to consider when evaluating water contamination. BOD5 is the amount of oxygen consumed in five days by microorganisms that oxidize biodegradable organic materials in an aerobic biochemical manner. The primary objective of this effort is to use microbial fuel cells (MFCs) to shorten the time required for BOD5 measurements. We created a regression artificial neural network (AI), and the predictions we obtained for BOD5 measurements were taken over 6 – 24 hours with an average error of just 7%. The outcomes demonstrated by our AI MFC/BES BOD5 sensor’s viability for use in real-world scenarios.

Objective: The aim of this study was to provide a new machine learning method to determine temporal events and inner-cycle parameters (e.g., cycle, poles and skis contact and swing time) in cross-country roller ski skating on the field, using a single deported inertial measurement unit (IMU). Methods: The developed method is based on long short-term memory neural networks to detect poles and skis initial and final contact with the ground during the cyclic movements. Eleven athletes skied four laps of 2.5 km at low and high intensity using skis with two different rolling coefficients. They were equipped with IMUs attached to the upper back, lower back and to the sternum. Data from force insoles and force poles were used as reference system. Results: The IMU placed on the upper back provided the best results, as the LSTM network was able to determine the temporal events with an accuracy ranging from 49 to 55 ms and the corresponding inner-cycles parameters were calculated with a precision of 63 to 68 ms. The method detected 95% of the events for the poles and 87% of the events for the skis. Conclusion: The proposed LSTM method provides a promising tool for assessing temporal events and inner-cycle phases in roller ski skating showing the potential of using a deported IMU to estimate different spatio-temporal parameters of human locomotion.

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.

Objective A trained T1 class Convolutional Neural Network (CNN) model will be used to examine its ability to successfully identify motor imagery when fed pre-processed electroencephalography (EEG) data. In theory, and if the model has been trained accurately, it should be able to identify a class and label it accordingly. The CNN model will then be restored and used to try and identify the same class of motor imagery data using much smaller sampled data in an attempt to simulate live data. Approach PyCharm, a Python platform, will be used to house and process the CNN. The raw data used for the training of the CNN will be sourced from the PhysioBank website. The EEG signal data will then be pre-processed using Brainstorm software that is a toolbox used in conjunction with MATLAB. The sample data used to validate and test the trained CNN, will be also be extracted from Brainstorm but in a much smaller size compared to the training data which is comprised of thousands of images. The sample size would be comparable to a person wearing a Brain Computer Interface (BCI), offering approximately 20 seconds of motor imagery signal data. Results The raw EEG data was successfully extracted and pre-processed. The deep learning model was trained using the extracted image data along with their corresponding labels. After training, it was able to accurately identify the T1 class label at 100 percent. The python coding was then modified to restore the trained model and feed it test sample data in which it was found to recognise 6 out of 10 lines of T1 signal image data. The result suggested that the initial training of the model required a different, more varying approach, so that it would be able to detect varying sample signal image data. The outcome of which could mean that the model could be used in applications for multiple patients wearing the same BCI hardware to control a device or interface.

Stripe rust ​​(caused by Puccinia striiformis f. sp. tritici) is one of the most devastating diseases of wheat and causes large-scale epidemics and severe yield loss. Applying fungicides during early epidemic development is crucial to controlling the disease but is often challenged by resource-limited human visual scouting. Deep learning has the potential to process images and videos captured from affordable devices to empower high-throughput phenotyping for early detection of stripe rust for timely application of fungicides and improve control efficiency. Here, we developed RustNet, a neural network-based image classifier, for efficiently monitoring fields for stripe rust. RustNet was built on a ResNet-18 architecture pre-trained with ImageNet Large-Scale Visual Recognition Challenge (ILSVRC) dataset using transfer learning. RGB images and videos of multiple wheat fields with different wheat types (winter and spring wheat), conditions (irrigated and non-irrigated), and locations were acquired using smartphones or unmanned aerial vehicles near the canopy. A semi-automated image labeling approach was conducted to improve labeling efficiency by combining automated machine labeling and human correction. Cross-validations across multiple categories (sensor platforms, wheat types, and locations) achieved Area Under Curve from 0.72 to 0.87. Independent validation on a published dataset from Germany achieved accuracies ranging from 0.79 to 0.86. The visualization of the last convolutional layer of RustNet demonstrated the identification of pixels with stripe rust. RustNet is freely available at https://zzlab.net/RustNet.

This article presents new PEM fuel cell models with hexagonal and pentagonal designs. After observing cell performance improvement in these models, we optimized them. Inlet pressure and temperature were used as input parameters, and consumption and output power were the target parameters of the multi-objective optimization algorithm. Then we used artificial intelligence techniques, including deep neural networks and polynomial regression, to model the data. Next, we employed the RSM (Response Surface Method) method to derive the target functions. Furthermore, we applied the NSGA-II multi-objective genetic algorithm to optimize the targets. Compared to the base model (Cubic), the optimized Pentagonal and Hexagonal models averagely increase the output current density by 21.819\% and 39.931\%, respectively.