Neural Network Technology Research Articles

Recent Advances in Machine Learning Assisted Hydrogel Flexible Sensing

Hydrogel flexible sensors are widely used in wearable devices, health care, intelligent robots and other fields due to their excellent flexibility, biocompatibility and high sensitivity. With the development of single sensor to multi‐channel and multi‐mode sensor network, the sensor data also presents the characteristics of multi‐dimension, complex and massive. Traditional data analysis methods can no longer meet the data analysis requirements of hydrogel flexible sensor networks. The introduction of machine learning (ML) technology optimizes the process of data analysis. With the continuous development of multi‐layer neural network technology and the improvement of computer performance, deep learning (DL) algorithm is increasingly used to achieve higher efficiency and accuracy, provides a powerful tool for data analysis of hydrogel flexible sensor, and accelerates the intelligent process of hydrogel flexible sensor equipment. This paper introduces the classification of hydrogel flexible sensors and the working mechanism and common algorithms of ML, and summarizes the application of ML technology to assist hydrogel flexible sensors in data analysis in the fields of health care and information recognition. This review will provide inspiration and reference for integrating ML technology into the field of hydrogel flexible sensors.

Enhancing Knowledge Graph Embedding with Hierarchical Self-Attention and Graph Neural Network Techniques for Drug-Drug Interaction Prediction in Virtual Reality Environments

In biomedicine, the critical task is to decode Drug–Drug Interactions (DDIs) from complex biomedical texts. The scientific community employs Knowledge Graph Embedding (KGE) methods, enhanced with advanced neural network technologies, including capsule networks. However, existing methodologies primarily focus on the structural details of individual entities or relations within Biomedical Knowledge Graphs (BioKGs), overlooking the overall structural context of BioKGs, molecular structures, positional features of drug pairs, and their critical Relational Mapping Properties. To tackle the challenges identified, this study presents HSTrHouse an innovative hierarchical self-attention BioKGs embedding framework. This architecture integrates self-attention mechanisms with advanced neural network technologies, including Convolutional Neural Network (CNN) and Graph Neural Network (GNN), for enhanced computational modeling in biomedical contexts. The model bifurcates the BioKGs into entity and relation layers for structural analysis. It employs self-attention across these layers, utilizing PubMedBERT and CNN for position feature extraction, and a GNN for drug pair molecular structure analysis. Then, we connect the position and molecular structure features to integrate them into the self-attention calculation of entity and relation. After that, the output of the self-attention layer is combined with the connected vectors of the position feature and molecular structure feature to obtain the final representation vector, and finally, to model the Relational Mapping Properties (RMPs), the representation vector is embedded into the complex vector space using Householder projections to obtain the BioKGs model. The paper validates HSTrHouse’s efficacy by comparing it with advanced models on three standard BioKGs for DDIs research.

AI- POWERED BASED BLIND AND VISUALLY IMPAIRED SYSTEM FOR SMART GLASS

Visual impairment causes serious problems in daily life and mobility for millions of people worldwide. Against this background, smart assistive technology is gaining momentum to improve the independence and quality of life of the blind and visually impaired. They have difficulty in daily life because they cannot detect the problems around them, and one of their biggest problems is identifying people. Besides automation, there are many unexplored uses of search engines. This project includes an app that uses search to help blind people see things in front of them safely, and a facial recognition system with visual feedback that can help visually impaired people see familiar and unfamiliar people. The speaker will provide them with sound support. In this work, we use deep learning-based Regional Convolutional Neural Network (Faster R-CNN) to detect and identify people and objects in the environment. Faster local convolutional neural network technology processes and distributes images captured by cameras. The operator takes the image as input. Therefore, this model helps visually impaired people in a simpler way than a white cane. Keywords—Visuallmpairment,Objectdetection,Faster Region Convolutional Neural Network.

Application of image rendering based on improved neural networks and sensors in landscape design

Landscape planning and design is an indispensable part of modern urban construction. In landscape design, rendering technology can transform designers' imagination into actual images, allowing people to better understand the design intent. In order to solve the problem of insufficient precision in image rendering in landscape design, this paper proposes a landscape design image rendering technology based on improved neural network technology. This technology combines stereo matching algorithms, which evaluate the corrected stereo images to calculate the differences between the real scene and the underlying model. The stereo matching algorithm is based on the basic model and obtains accurate depth information from a wide range image pair, making the rendered landscape image more realistic and precise. The experiment demonstrates the effectiveness of the landscape design image rendering system through two important aspects: processing speed and parallax accuracy. The development of image rendering technology in landscape design cannot be separated from the mutual promotion of theory and practice. The technology proposed in this article has certain practical value and hopes to make more positive contributions to urban construction and cultural inheritance.

Neural Technologies for Objects Classification with Mobile Applications

This paper is related to the study of the features of the neural technologies’ application, in particular, ResNet neural networks for the classification of objects in photographs. The work aims to increase the accuracy of recognition and classification of objects in photographs by using various models of the ResNet neural network. The paper analyzes the features of the application of the corresponding models in comparison with other architectures of deep neural networks and evaluates their efficiency and accuracy in the classification of objects in photographs. The process of data formation for training neural networks, their processing and sorting is described. A web application and a mobile application for recognizing and classifying objects in a photo were also developed. A system for classifying objects, in particular airplanes in photographs, was developed using neural network technologies. It gives a recognition and classification accuracy of about 95%. Research results of ResNet models are of great practical importance, as they can improve the classification accuracy of various images. Features of ResNet, such as the use of skip connections or residual connections, make it effective in the relevant tasks. The results of the study will help to implement ResNet in various fields, including medicine, automatic pattern recognition and other areas where the classification of objects in photographs is an important task.

Research on Predicting Welding Deformation in Automated Laser Welding Processes with an Enhanced DEWOA-BP Algorithm

Research on Predicting Welding Deformation in Automated Laser Welding Processes with an Enhanced DEWOA-BP Algorithm

Research on a Real-Time Prediction Method of Hull Girder Loads Based on Different Recurrent Neural Network Models

Real-time prediction of hull girder loads is of great significance for the safety of ship structures. Some scholars have used neural network technology to investigate hull girder load real-time prediction methods based on motion monitoring data. With the development of deep learning technology, a variety of recurrent neural networks have been proposed; however, there is still a lack of systematic comparative analysis on the prediction performance of different networks. In addition, the real motion monitoring data inevitably contains noise, and the effect of data noise has not been fully considered in previous studies. In this paper, four different recurrent neural network models are comparatively investigated, and the effect of different levels of noise on the prediction accuracy of various load components is systematically analyzed. It is found that the GRU network is suitable for predicting the torsional moment and horizontal bending moment, and the LSTM network is suitable for predicting the vertical bending moment. Although filtering has been applied to the original noise data, the prediction accuracy still decreased as the noise level increased. The prediction accuracy of the vertical bending moment and horizontal bending moment is higher than that of the torsional moment.

Моделирование системы управления отоплением дома с применением нейросетевых технологий

The aim of the work is to develop methods to improve the characteristics of existing heating systems through the use of a new control algorithm based on the parallel principle of operation and using neural network technologies. The proposed control method provides for the analysis by a special computer program of feature matrices consisting of both control actions and resulting maps. The computer, which knows in advance that different images correspond to different modes, selects the appropriate mode based on the image received at the program input. This mode can be selected in such a way that there will be minimal energy consumption while providing the conditions that the user needs. An application has been developed that uses a neural network algorithm, with the ability to change settings. Sets of images have been prepared indicating class membership by characteristic features (datasets), on which the neural network was trained and tested. More than 600 experiments with a different number of learning epochs have been performed. The influence of the number of epochs on the average percentage of image recognition of building temperature regimes has been studied. To demonstrate the possibility of implementing the proposed algorithm, a control system architecture and a distributed application consisting of a main control program with a human-machine interface and a remote web application with a human-machine interface and a microservice with an access point were developed. A mock-up of the building was made with a connected room temperature control system. The experiments carried out have confirmed the effectiveness of the proposed control method. This technology provides for the possibility of switching from managing a single building to managing an array of buildings without changing the management settings of individual buildings. Using this method will increase the convenience of management, the comfort of people's living conditions and reduce energy consumption for heating.

BTR: a bioinformatics tool recommendation system.

The rapid expansion of Bioinformatics research has led to a proliferation of computational tools for scientific analysis pipelines. However, constructing these pipelines is a demanding task, requiring extensive domain knowledge and careful consideration. As the Bioinformatics landscape evolves, researchers, both novice and expert, may feel overwhelmed in unfamiliar fields, potentially leading to the selection of unsuitable tools during workflow development. In this paper, we introduce the Bioinformatics Tool Recommendation system (BTR), a deep learning model designed to recommend suitable tools for a given workflow-in-progress. BTR leverages recent advances in graph neural network technology, representing the workflow as a graph to capture essential context. Natural language processing techniques enhance tool recommendations by analyzing associated tool descriptions. Experiments demonstrate that BTR outperforms the existing Galaxy tool recommendation system, showcasing its potential to streamline scientific workflow construction. The Python source code is available at https://github.com/ryangreenj/bioinformatics_tool_recommendation.

Unsupervised Social Bot Detection via Structural Information Theory

Research on social bot detection plays a crucial role in maintaining the order and reliability of information dissemination while increasing trust in social interactions. The current mainstream social bot detection models rely on black-box neural network technology, e.g., Graph Neural Network, Transformer, etc., which lacks interpretability. In this work, we present UnDBot, a novel unsupervised, interpretable, yet effective and practical framework for detecting social bots. This framework is built upon structural information theory. We begin by designing three social relationship metrics that capture various aspects of social bot behaviors: Posting Type Distribution , Posting Influence , and Follow-to-follower Ratio . Three new relationships are utilized to construct a new, unified, and weighted social multi-relational graph, aiming to model the relevance of social user behaviors and discover long-distance correlations between users. Second, we introduce a novel method for optimizing heterogeneous structural entropy. This method involves the personalized aggregation of edge information from the social multi-relational graph to generate a two-dimensional encoding tree. The heterogeneous structural entropy facilitates decoding of the substantial structure of the social bots network and enables hierarchical clustering of social bots. Thirdly, a new community labeling method is presented to distinguish social bot communities by computing the user’s stationary distribution, measuring user contributions to network structure, and counting the intensity of user aggregation within the community. Compared with ten representative social bot detection approaches, comprehensive experiments demonstrate the advantages of effectiveness and interpretability of UnDBot on four real social network datasets.

Changing Trends in Teaching Computer Vision at Ukrainian Universities in the Age of Artificial Intelligence

In recent years, professional education in computer vision and artificial intelligence (AI) has witnessed transformative changes, a process significantly accelerated by the COVID-19 pandemic, which catalyzed the shift towards online learning formats. In Ukraine, as in many other countries, universities have rapidly adapted their curricula to align with these new realities. The onset of martial law in Ukraine further entrenched online learning as the dominant mode of education. Additionally, there has been a noticeable shift in research activities towards projects tailored to the defense sector, unfolding against the backdrop of significant global advancements in AI for computer vision. This article explores the impact of these shifts and the “Artificial Intelligence Development Concept” introduced by the Ministry of Digital Transformation of Ukraine on the quality and methodologies of computer vision education. It scrutinizes specific training courses in this domain, along with the literature used in the educational process. The exploration includes how deep neural network technologies such as TensorFlow, PyTorch, OpenCV, and Python libraries are being integrated into computer vision curricula, reflecting the latest advancements in the field. It also examines the needs of Ukrainian startups and companies specializing in computer vision projects, analyzing how curricula are evolving to meet the specific requirements of modern software developers. Furthermore, the article proposes directions for the further development of online education in computer vision teaching in Ukraine.

Intelligent Power Management of Electric vehicle with onboard PV by ANN-based Model Predictive Control

This paper presents a novel configuration of Hybrid Electric Vehicle (HEV) with fuel cell as primary source and an on-board Photovoltaic (PV) array as secondary source. It incorporates an advanced Model Predictive Control (MPC) system that utilizes Artificial Neural Network (ANN) technology to control induction motor of electric vehicle. The system has three main parts: a fuel cell, an electrolyzer, and a PV array. Each part has a different job in supplying power to the vehicle. The PV array functions to supplement the fuel cell by providing power when ample irradiation is available. When the car isn't being used, the extra power from the solar panels goes to the electrolyzer. There, it turns into chemical energy, making hydrogen. This hydrogen is stored in the onboard hydrogen tank for future use. To fulfil the voltage needs of both the motor and energy sources, a quadratic bidirectional buck-boost converter (QBBC) is utilized. This converter efficiently manages the voltage output while ensuring compatibility with various energy sources. The system undergoes comprehensive analysis under different irradiance and speed conditions to evaluate its performance and efficiency. To make the solar power system more efficient and cost-effective, a Maximum Power Point Tracking (MPPT) algorithm is used. This algorithm helps get the maximum power from the PV system. An improved incremental conductance algorithm is used for this, making sure it accurately finds the best power point. The control strategy incorporates both outer voltage and inner current control, to effectively control the DC output voltage of the QBBC. The vehicle's drive system employs an indirect vector-controlled induction motor. To regulate the motor's speed effectively, an advanced ANN-MPC system is used, providing precise control and high efficiency. This paper introduces a novel approach for predictive torque control of AC machines, eliminating the need for weighting factors. An ANN estimator is used to accurately expect the motor speed, which improves the overall performance of the control system. By using MATLAB/SIMULINK to test how well the proposed Electric Vehicle (EV) setup works and make sure it's effective in real-life situations. Through this comprehensive approach, the paper contributes to advancing the development of efficient and intelligent electric vehicle systems..

Computer-aided mobility solutions: Machine learning innovations to secure smart urban transportation

This study introduces a groundbreaking hybrid deep learning Intrusion Detection System (IDS) tailored to improve the security of Electric Vehicle (EV) charging processes in the context of computer-aided mobility solutions for smart urban transportation. The proposed system ingeniously integrates XGBoost and Convolutional Neural Network (CNN) technologies, leveraging their combined strengths to fortify the resilience of EV charging infrastructure. In response to the demand for transparency in decision-making, our approach incorporates Explainable AI Techniques, specifically employing Shapley Additive exPlanations (SHAP) values. This not only enables the IDS to identify anomalous behavior but also provides clear insights into the features influencing the detection process. The transparency achieved is pivotal for fostering stakeholder trust, ensuring regulatory compliance, and facilitating well-informed decision-making within the dynamic landscape of smart urban transportation. A sophisticated mechanism is implemented to enable the model to autonomously adjust and learn from evolving charging patterns over time. This adaptive capability ensures the IDS remains effective in the face of changing charging behaviors, contributing to the resilience of the overall smart urban transportation system. This research not only propels the field of cybersecurity in smart urban transportation but also establishes a robust framework for the development of intelligent systems. These systems play a critical role in securing essential processes while providing interpretable insights for stakeholders in the realm of computer-aided mobility solutions.

Automated Classification of Abdominal Ultrasound lmages of the Pancreas Based on the Spectral Representation of the Border’s Contours

The purpose of the research is to develop a methodology for classifying complexly structured halftone images based on a multimodal approach using methods of morphological analysis, spectral analysis and neural network modeling.Methods. А method for classifying the contours of the boundaries of segments of a complexly structured image is described. Тhe method is based on the fact that in chronic diseases of the pancreas, there is a violation of the integrity of the contour of its border and its waviness increases due to retractions and bulges caused by an alterative inflammatory process. Тhe method includes the stages of normalization of ultrasound images and image segmentation with the selection of the contour of the object of interest. Тo classify the contour of a segment boundary, it is proposed to use Fourier analysis and neural network technologies. Тhe method is illustrated using the example of classifying the contour of the border of the pancreas on its transcutaneous acoustic image.Results. Еxperimental studies of the proposed methods and means for classifying medical risk were carried out on diagnostic tasks according to the following classes: "chronic pancreatitis" – "without pathology". For experimental studies, video sequences of ultrasound images of the pancreas provided by an endoscopist were used. Тhe purpose of the experimental studies was to analyze the classification quality indicators of image classifiers with class segments "Chronic pancreatitis" and "Without pathology". Тhe training sample of video images (frames of video sequences) included 200 examples, one hundred from each class. Тhe quality indicator "Sensitivity" of classification for two classes is 85,7%, the indicator "Specificity" is 87,1%.Тhe use of the contour analysis method in classifiers of ultrasound images of the pancreas opens up new opportunities for accessible and objective diagnosis of pancreatic diseases, expanding the capabilities of intelligent clinical decision support systems.

Research on Power Equipment Fault Diagnosis Based on Improved SVM Algorithm

Recently, Data collected from safety relays and circuit breakers has been used in fault diagnostics to identify power system components that are either failing or have tripped. A tiny percentage of heating zones may cause problems when identifying faults, although the chance of aberrant heat production in power equipment is modest. The power transformer is one of the most crucial components of any electrical system. Discovering any concealed defects in power equipment is essential to ensuring a continuous supply of electrical power for customers. The challenging characteristic of such fault diagnosis in power equipment is the greater potential for harm and a great deal of time invested in testing and training. Hence, in this research, Smart Support Vector Machine enabled Artificial Neural Network (SVM-ANN) technologies have been improved for fault diagnosis in power equipment for industry. The Genetic Algorithm (GA) process the power equipment for industrial data processing and monitoring in that fault diagnosis. The classification of strategies used and their relationships to power equipment processes to identify significant trends and research problems related to intelligent fault diagnosis systems for industry. The experimental analysis of SVM-ANN outperforms fault diagnosis using the power equipment industry in terms of performance, accuracy, prediction ratio, efficiency, and error rate.

Fusing Deep Immuno Net and Random Forest to Improve the Detection of Dengue Fever: A Computational Strategy for Prompt and Precise Diagnosis

Dengue fever is a worldwide health issue caused by the virus. Effective medical management and avoidance of severe consequences need fast and precise Dengue detection. This research uses advanced image pre-processing, deep learning methods using convolutional neural networks (CNNs), and feature selection to improve Dengue diagnosis. Our first study focuses on improving input images using cutting-edge image enhancement methods, including Adaptive Contrast Enhancement using Histogram Equalization (ACE-HE)during pre-processing. This improves visual data for processing, enabling Dengue identification. Feature extraction, using a CNN architecture optimized for Dengue detection, is our core technique. We also use current neural network designs like EfficientNet and transformers to extract subtle characteristics needed for effective diagnosis. Our approach uses computational evolutionary algorithms and neural structure searches to identify the most relevant features from the large pool of retrieved information. These characteristics are gradually fused to provide a complete depiction of Dengue's complicated patterns. Dengue detection is completed using high-performance classifiers including Random Forest variations and ensemble approaches. Our architecture is adaptable and resilient, achieving above 97.92% BCCD dataset classification accuracy. This technology represents a major advance in Dengue diagnosis and meets the worldwide requirement for rapid and accurate infectious illness identification. Our strategy uses the latest computer vision and neural network technology to produce dependable and effective tools for rapid and accurate Dengue diagnosis, addressing a crucial worldwide healthcare issue.

Application of deep learning model incorporating domain knowledge in international migration forecasting

PurposeForecasting population movement trends is crucial for implementing effective policies to regulate labor force growth and understand demographic changes. Combining migration theory expertise and neural network technology can bring a fresh perspective to international migration forecasting research.Design/methodology/approachThis study proposes a conditional generative adversarial neural network model incorporating the migration knowledge – conditional generative adversarial network (MK-CGAN). By using the migration knowledge to design the parameters, MK-CGAN can effectively address the limited data problem, thereby enhancing the accuracy of migration forecasts.FindingsThe model was tested by forecasting migration flows between different countries and had good generalizability and validity. The results are robust as the proposed solutions can achieve lesser mean absolute error, mean squared error, root mean square error, mean absolute percentage error and R2 values, reaching 0.9855 compared to long short-term memory (LSTM), gated recurrent unit, generative adversarial network (GAN) and the traditional gravity model.Originality/valueThis study is significant because it demonstrates a highly effective technique for predicting international migration using conditional GANs. By incorporating migration knowledge into our models, we can achieve prediction accuracy, gaining valuable insights into the differences between various model characteristics. We used SHapley Additive exPlanations to enhance our understanding of these differences and provide clear and concise explanations for our model predictions. The results demonstrated the theoretical significance and practical value of the MK-CGAN model in predicting international migration.

A convolutional neural network prediction model for aviation nitrogen oxides emissions throughout all flight phases

In recent years, there has been an increasing amount of research on nitrogen oxides (NOx) emissions, and the environmental impact of aviation NOx emissions at cruising altitudes has received widespread attention. NOx may play a crucial role in altering the composition of the atmosphere, particularly regarding ozone formation in the upper troposphere. At present, the ground emission database based on the landing and takeoff (LTO) cycle is more comprehensive, while high-altitude emission data is scarce due to the prohibitively high cost and the inevitable measurement uncertainty associated with in-flight sampling. Therefore, it is necessary to establish a comprehensive NOx emission database for the entire flight envelope, encompassing both ground and cruise phases. This will enable a thorough assessment of the impact of aviation NOx emissions on climate and air quality. In this study, a prediction model has been developed via convolutional neural network (CNN) technology. This model can predict the ground and cruise NOx emission index for turbofan engines and mixed turbofan engines fueled by either conventional aviation kerosene or sustainable aviation fuels (SAFs). The model utilizes data from the engine emission database (EEDB) released by the International Civil Aviation Organization (ICAO) and results obtained from several in-situ emission measurements conducted during ground and cruise phases. The model has been validated by comparing measured and predicted data, and the results demonstrate its high prediction accuracy for both the ground (R2 > 0.95) and cruise phases (R2 > 0.9). This surpasses traditional prediction models that rely on fuel flow rate, such as the Boeing Fuel Flow Method 2 (BFFM2). Furthermore, the model can predict NOx emissions from aircrafts burning SAFs with satisfactory accuracy, facilitating the development of a more complete and accurate aviation NOx emission inventory, which can serve as a basis for aviation environmental and climatic research. SynopsisThe utilization of the ANOEPM-CNN offers a foundation for establishing more precise emission inventories, thereby reducing inaccuracies in assessing the impact of aviation NOx emissions on climate and air quality.

Correlative Analysis and Prediction of Physical Education Data Via Machine Learning: A Case Study on Grade Evaluation Method in University

Physical education is an important way to strengthen the physical quality of students in the university. Sport performance is an important criterion for judging the physical fitness of college students. But during the process of calculating and collecting the comprehensive score for each student, teachers will use different equipment or various grading formulas to evaluate the grade for many years. Thus, referenced score will lose their values if the measurement data for each student in different years are not unified and sometimes with the subjective factor because of the manual calculations. This article analyzes the details of physical fitness test results and the relationship with the comprehensive score. It is discussed that the comprehensive score can provide serious help for teachers to know students’ physical quality level and make reasonable teaching program to each student according to their personal radar chat. Therefore, the artificial influence is excluded, and the prediction model is designed by using the principal component analysis method and the back propagation (BP) neural network technology. The predictive model allows students to evaluate their own physical test score in advance, get a preliminary understanding of their physical fitness. Comparisons are made between the applications of this model in different years and errors are discussed to verify the accuracy. The results indicate that the comprehensive score prediction model supplies one effective approach to unify the scoring standards and improve the computation efficiency in physical education.

Integration of chatbots into foreign language learning

Innovative technologies are a part and parcel of modern education. Being integrated in the education process, neural network technologies are of particular interest now. Chatbots are used for different purposes: to gain and consolidate knowledge, to test knowledge, to study foreign languages, to organize the education process, etc. The use of generative AI contributes to education efficiency, improves skills and students’ motivation. One of the most popular and promising generative neural networks is ChatGPT. The use of this chatbot got mixed reviews, and this article examines the possibilities of using chatbots in education. We attempt to use ChatGPT as a support tool for compiling a training manual on developing listening skills for EFL (English as a foreign language) students. We analyze the advantages and disadvantages of using ChatGPT for compiling teaching materials and outline recommendations on how to use this tool efficiently. We found that, in general, the use of ChatGPT optimizes the process of preparing training materials for intermediate-level students. However, it is important to take into account a number of features such as the language register, the genre and text type, semantic breakdown of the text, the risk of neural network making an error, etc. The article concludes that the chatbot, if used correctly, can become an effective virtual assistant for an EFL teacher.