Neural Network Analysis Method Research Articles

A New Approach for Predicting Strength Based on Temperature-Time History Using Two-Parameter Maturity ANN Models.

One widely used method to predict concrete strength development based on temperature variations during curing is the equivalent maturity time (te) method. This method uses the activation energy (Ea) as its key parameter, which reflects the cement's sensitivity to temperature. However, research shows that the Ea value varies depending on factors such as cement type, water/cement ratio, temperature, and additives. The permanent subject of discussion is the question of what value of the Ea parameter should be assumed. In this paper, a new approach is proposed by using a neural network analysis method to develop a strength-temperature history model. It was assumed that the ANN-fc% = f(Q, E, T, t) model would have 4 inputs: hydration heat (Q), activation energy (Ea), temperature (T), and time (t). The research was conducted on mortars using 6 cements, at curing temperatures ranging from 5 to 35 °C, assessing strength over a 90 day period. The results showed that the ANN analysis method allows for estimating the relative compressive strength with sufficient accuracy. Analysis of the input nodes indicated that Q influences early strength gain, while Ea affects later strength development. The application of the ANN model for calculating strength based on temperature changes during maturation was illustrated.

Dynamic behavior of high-speed maglev guideway girders with quadratically-varied cross-section

This study focuses on the numerical investigation on the coupling vibration responses of high-speed maglev vehicle-guideway system incorporated with guideway girders with quadratically-varied cross-section and the new generation high-speed maglev trains. Finite element model for the coupling vibration responses analysis of maglev vehicle-guideway system was established and validated against the existing field test results. With the verified model and program, a total of 5292 numerical models were included to comprehensively investigate the influences of cross-sectional geometries and span length of the guideway girder as well as the vehicle speed on the coupling vibration responses of high-speed maglev vehicle-guideway system. The coupling vibration responses, including impact coefficient, maximum suspension gap and maximum displacement of bogie as well as maximum displacement of the car-body, were analyzed. Recommendation on the dynamic performance indicator was proposed for the guideway girders with quadratically-varied cross-section. Prediction models regarding the impact coefficient and mid-span maximum acceleration of guideway girders with quadratically-varied cross-section were developed based on the Artificial Neural Network analysis method.

Research on the correlation between roughness parameters and contact stress on tooth surfaces and its dominant characteristics

The contact performance of the tooth surface is a crucial factor affecting the operational lifespan of gears. There is a significant correlation between tooth surface topography and contact stress, as well as fatigue performance. To investigate the influence of different topography parameters on tooth surface contact stress, this paper performs analytical calculations on asperity contacts of rough tooth surfaces based on reconstruction modeling. Through efficient calculations, sufficient correlation analysis data is obtained. The neural network analysis method is used to measure the degree of influence of roughness parameters on tooth surface contact performance. Research on the correlation between micro-topography parameters and maximum contact stress on tooth surfaces, as well as regression analysis, is conducted. The results show that: (1) Considering rough surfaces, the tooth surface contact stress field exhibits two stress peaks in the near-surface layer and sub-surface layer, with a maximum Mises stress increase of up to 30 % compared to a smoother surface. (2) In the case of low roughness (Sa < 0.2 μm), the amplitude of the maximum Mises stress peak in the near-surface layer is close to that in the sub-surface layer. However, as the roughness amplitude increases, the amplitude of the near-surface stress peak increases sharply. (3) Ranking the roughness parameters based on their influence on contact stress, the main parameters characterizing contact performance are identified as Spk, Vmp, and Sq. The peak characteristics of the surface are the decisive factors affecting fatigue extremes, which are stronger than the traditional surface roughness parameter Sq. This paper establishes a direct correlation model between tooth surface topography parameters and contact stress, providing a foundation for simplifying stress calculations considering complex topographies and offering new insights for fatigue-resistant design of tooth surfaces.

Sistem Presensi Pendeteksi Wajah menggunakan Metode Modified Region Convolutional Neural Network dan PCA

Attendance is an activity to collect data to find out the number of attendances at an event or activity. In order to increase effectiveness and time efficiency is important, it is necessary to have an artificial intelligence-based presence by means of face detection. This study aims to create a face detection attendance system using modified region convolutional neural network (MR-CNN) and principal component analysis (PCA) methods. This type of research is development research by applying the concept of digital image based on face recognition. This research applies a method in deep learning, namely MR-CNN through camera media to take images and the extraction process using the PCA method to reduce image resolution. Images that include various individuals who want to be recognized are stored and then used as datasets. From the dataset, the MR-CNN model was trained with training data. Our findings are in the form of a web-based facial short-attendance system. Where are the calculation results in this system has an accuracy value of 96.1%, so it can be used well for facial identification, and can be used at SMK Bhakti Anindya.

Exploring the recycled water acceptance based on the technological perspective of UTAUT2: a hybrid analytical approach.

The development of advanced sewage technologies empowers the industry to produce high-quality recycled water, which greatly influences human's life and health. Thus, this study investigates the mechanism of individuals' adoption of recycled water from the technology adoption perspective. Employing the mixed method of structural equation modeling and artificial neural network analysis, we examined a research model developed from the extended Unified Theory of Acceptance and Use of Technology (UTAUT2) framework. To examine the research model, this study employs a leading web-survey company (Sojump) to collect 308 valid samples from the residents in mainland China. The structural equation modeling results verified the associations between the six predictors (performance expectancy, effort expectancy, social influence, facilitating conditions, environmental motivation, and price value), individuals' cognitive and emotional attitudes, and acceptance intention. The artificial neural network analysis validates and complements the structural equation modeling results by unveiling the importance rank of the significant determinants of the acceptance decisions. The study provides theoretical implications for recycled water research and useful insights for practitioners and policymakers to reduce the environmental hazards of water scarcity.

Analysis of Local Characteristic Tourism based on Gray Prediction and BP Neural Networks

This study explores the strategy of utilizing online platforms to promote the recovery and development of the tourism industry against the background of the impact of the 2020 COVID-19 on the tourism industry. The development trend of Harbin's characteristic tour-ism industry was studied through college students' survey data, combined with multiple regression, gray prediction and BP neural network analysis methods. The results of the study show that combining network publicity with special tourism programs and attracting young people groups is the key, and suggestions for sustainable development are put forward. By helping the inheritance and innovation of local traditional culture, this study aims to promote the innovative development of tourism. This research framework provides theoretical and practical guidance for promoting the development of local tourism.

Financial priorities and the growth border of the Russian economy

Subject. The article investigates the role of foreign exchange reserves in the development of the Russian economy, the securities market development, and the banking system stability. Objectives. The aim is to review the monetary policy options under international sanctions strengthening. Methods. I apply methods of statistical, neural network, and cluster analysis. Results. The study established that it is impossible to reduce currency risks by issuing shares listed in the domestic market in foreign currency. The preferred directions of Russia's monetary policy are as follows: strengthening the country's international investment position, increasing the volume of the Federal bond market, and increasing the issue of shares listed in the domestic market in rubles. Conclusions. The growth boundary of the Russian economy is determined by such parameters as the amount of funds in escrow accounts of individuals, the size of insurance and pension reserves.

Meta-DHGNN: method for CRS-related cytokines analysis in CAR-T therapy based on meta-learning directed heterogeneous graph neural network.

Chimeric antigen receptor T-cell (CAR-T) immunotherapy, a novel approach for treating blood cancer, is associated with the production of cytokine release syndrome (CRS), which poses significant safety concerns for patients. Currently, there is limited knowledge regarding CRS-related cytokines and the intricate relationship between cytokines and cells. Therefore, it is imperative to explore a reliable and efficient computational method to identify cytokines associated with CRS. In this study, we propose Meta-DHGNN, a directed and heterogeneous graph neural network analysis method based on meta-learning. The proposed method integrates both directed and heterogeneous algorithms, while the meta-learning module effectively addresses the issue of limited data availability. This approach enables comprehensive analysis of the cytokine network and accurate prediction of CRS-related cytokines. Firstly, to tackle the challenge posed by small datasets, a pre-training phase is conducted using the meta-learning module. Consequently, the directed algorithm constructs an adjacency matrix that accurately captures potential relationships in a more realistic manner. Ultimately, the heterogeneous algorithm employs meta-photographs and multi-head attention mechanisms to enhance the realism and accuracy of predicting cytokine information associated with positive labels. Our experimental verification on the dataset demonstrates that Meta-DHGNN achieves favorable outcomes. Furthermore, based on the predicted results, we have explored the multifaceted formation mechanism of CRS in CAR-T therapy from various perspectives and identified several cytokines, such as IFNG (IFN-γ), IFNA1, IFNB1, IFNA13, IFNA2, IFNAR1, IFNAR2, IFNGR1 and IFNGR2 that have been relatively overlooked in previous studies but potentially play pivotal roles. The significance of Meta-DHGNN lies in its ability to analyze directed and heterogeneous networks in biology effectively while also facilitating CRS risk prediction in CAR-T therapy.

Assessment of the bone biological age using multiple regression and neural network analysis

Background. Nowadays, many studies confirm the increased risk of osteoporosis in the elderly. Strategies for optimizing diagnosis include a combined use of various methods, so calculating bone biological age (BA) can also be important for disease prediction. Recently, some new methodological approaches to BA calculation with the mathematical methods implementation were proposed. However, studies utilizing advanced approaches, particularly neural network (NN) in Ukraine, are limited. The purpose of this study was to develop a bone BA model and compare its accuracy using stepwise multiple regression (MLR) and NN analysis methods. Materials and methods. In a single-center cohort study, we retrospectively analyzed the data of 3,071 healthy women aged 40–90 years. The analysis of the study indices included the anthropometric parameters, dual-energy X-ray absorptiometry indices (DXA), and the parameters of the 10-year probability of major osteoporotic fractures (FRAX). For the development of bone BA models, MLR methods were used and the construction of the NN model was performed using a multilayer perceptron model. Results. As a result of the calculation, we received the MLR formula for bone BA determining bone mineral density (BMD) of lumbar spine and radius, minimal femoral and hip BMD as well as Trabecular Bone Score, and FRAX. The MLR equation allowed to calculation of the bone BA with an error of less than 4.9 years for study and control groups and demonstrated the high connection between calculated and chronological (ChrA) ages (R = 0.77; p &lt; 0.00001). The use of NN analysis showed the best results using 6 input variables and 1 internal layer of 7 neurons. The assessment of the connection between BA and ChrA demonstrated a high coefficient of correlation (R = 0.88; p &lt; 0.000001) with an average error of age calculation of less than 3.7 years for the study and control groups. Conclusions. A comparison of the accuracy of both models in bone BA estimation revealed a significant advantage of the deep learning NN, however, the use trained NN model requires specialized software, whereas the MLR formula can be used without additional costs.

A Study of Factors Influencing Happiness in Korea: Topic Modelling and Neural Network Analysis

The aim of this study is to derive the important factors that influence levels of happiness in Korea, and to identify which factors are particularly important among these influencing factors. To achieve this goal, topic modelling analysis, machine learning analysis and neural network analysis methods were utilized. The Netminer 4.5 program was used for topic modelling analysis and machine learning analysis, and SPSS MODELER 18 was used to perform neural network analysis. Two types of analysis data were used in this study. The first consisted of 1 000 papers relating to happiness published in academic journals managed by the Springer publishing company, which were used to derive happiness-influencing factors. The second consisted of a survey conducted in 2020 by the Community Well-being Center of the Graduate School of Public Administration at Seoul National University in Korea. A total of 16 655 people responded to this survey. The analysis results of the study are as follows. Important variables that affect the level of happiness of Korean residents are: family life, social status, income, health, and perceptions of inequality. Analysis using neural network analysis of the most important factors influencing happiness showed that satisfaction with family life had the most important influence. This suggests that policies that can improve the quality of family life, such as family-friendly work environments, childcare support, and domestic violence prevention and response programmes, will become important in the future

Aplicação de modelagem de tópicos e análise de redes neurais para analisar a satisfação com a vida

Abtract This study aims analyze the important influencing factors that affect the life satisfaction of Koreans, and to identify the relative importance of these factors. For this purpose, we utilize academic papers on what influences life satisfaction, and questionnaire data from the survey on social integration conducted annually by the Korean Government. A topic modelling analysis method was used to derive important influencing factors, and a neural network analysis method, one of the machine learning methods, was used to analyze the relative importance of influencing factors. The analysis showed that the factor that had the greatest impact on Koreans’ life satisfaction was satisfaction with work. Other factors included self-esteem, level of worry and anxiety, and level of satisfaction with health status. The study used methods such as topic modeling and neural network analysis to derive the main factors affecting life satisfaction and analyze the relative importance of the factor involved. The study results suggest that in recognition of the importance of job satisfaction, future research should be expanded, and that the Korean Government should introduce various policies to increase job satisfaction.

Sustainable Development Goals in Brazil in the context of the global environmental Agenda 21 (using the example of SDG 15)

The article examines the policy of one of the Latin American countries, Brazil, in the field of achieving SDG 15 (Protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss). Some aspects of achieving the Sustainable Development Goals have been identified. The research is based on a systemic approach using statistical and neural network analysis methods. The research revealed a contradiction expressed in the priorities of SDG 15 and Brazil’s institutional capabilities. For Brazil, the key problem is the deforestation, the reduction of biodiversity, and land degradation.

Bridging the gap: A comparative analysis of traditional and neural network regression methods for predicting university entrant performance in SUAST examination

In developing countries like the Philippines, access to free and high-quality tertiary education is crucial for better job opportunities. The State University Aptitude and Scholarship Test (SUAST) is used as a college admission examination by Davao Oriental State University (DOrSU). However, the passing rate for SUAST was only 54% for the academic year 2018-2023, and non-passers were still accepted due to policy changes, which undermine the purpose of the examination. This study aimed to identify the factors that influence the performance of university entrants in the SUAST examination using a researcher-made survey questionnaire administered online, utilizing both multiple-layer perceptron neural network (MLPNN) and multiple linear regression analysis (MLR) methods. A sample size of 359 was recommended, and the study found that family income, senior high school general weighted average (SHSGWA), library entry, intrinsic goal, openness and intellect, and behavioral reaction were significant predictors of SUAST exam scores. MLPNN analysis further identified library access and resources, family income, and academic self-belief as the most important predictors of SUAST exam scores, and MLPNN outperformed MLR. This study provides recommendations for DepEd and HEI’s to enhance the preparation and performance of students taking the SUAST exam, such as offering study materials and test-taking strategies, evaluating alternative admission tests, and reviewing the content validity of the questionnaire.

A Lightweight Neural Network for Spectroscopic Ellipsometry Analysis

AbstractEllipsometry is a widely used technique in thin film characterizations. To extract the optical properties of the films from the measured data, regression data fitting techniques have been developed that iteratively find a set of optical parameters that best fit the observations. However, this iterative process often faces challenges in converging to the correct solution, and it can be time‐consuming. To address these issues, an 8‐bit quantized lightweight method of neural network analysis for spectroscopic ellipsometry are proposed. This method features compact neural network modules to enhance speed and efficiency. The effectiveness of the approach through experimental verification is validated on a diverse range of material films, including metals, semiconductors, and dielectrics. The approach is fully automatic and lightweight, which offers a new perspective on balancing predictive accuracy with limited computational resources. This method holds the potential to achieve automatic, rapid, and high‐throughput optical characterization of films, facilitating real‐time quality monitoring for repeatable high‐precision film manufacturing.

Possibilities of monetary regulation in Russia

Subject. The article investigates monetary regulation in Russia. Objectives. The focus is on determining the possibilities of monetary regulation in Russia. Methods. The study draws on systems approach, using the methods of statistical, neural network, and cluster analysis. Results. The paper unveils peak values of intraday loans, other (lombard) loans, a strong direct relationship of intraday loans with other loans, and a decrease in overnight loans. It describes monetary inflationary risks, transition to a high level of standing deposit facilities (for overnight deposits) and one-week deposit auctions. I found significant dynamics of funds in escrow accounts, brokerage accounts and debt securities, demonstrated reaching a new level of debt relations of the non-financial sector, households, and non-profit organizations serving households. The study reveals the structural importance of the population in the context of the necessary link to GDP at constant prices in rubles. Conclusions. The real range of opportunities for monetary regulation in Russia for public authorities is associated with an increase in cash in circulation, growth in the level of standing deposit facilities, maintaining the excess of the debt of non-financial sector over the debt of households and non-profit organizations serving households, the dynamics of gross domestic product in constant prices in rubles and the population. The provisions of the study expand the competence of government members in the process of formulating a set of measures aimed at improving the effectiveness of monetary regulation.

Standard Setting with Artificial Neural Networks: TIMSS 2015 Mathematics Case

This study aims to demonstrate the optimal way to determine the cut-off score to be used to interpret the total scores obtained from an achievement test or scale using the Artificial Neural Networks method. To this end, the multiple-choice item responses in the Booklet-11 Mathematics subtest at the 8th grade level in the TIMSS 2015 Turkey sample dataset were used to determine the cut-off score for the achievement test. The item responses in the “Students Like Learning Mathematics Scale” in the TIMSS 2015 8th grade Mathematics Student Questionnaire were used to determine the cut-off score for the scale. The data were accessed from the TIMSS international database and the data were analyzed in MATLAB R2017b software. As a result of the study, the most appropriate cut-off score to be used for the evaluation of the total scores obtained from the TIMSS 2015 8th grade level Booklet-11 Mathematics subtest was determined as 45.5 out of 0-100 points with the Artificial Neural Network analysis method. The overall level of agreement between the cut-off score and the pass/fail classification based on 400 points, which is the lowest level of the TIMSS International Benchmark, was determined as 81%. The most appropriate cut-off score to be used for the evaluation of the scores obtained from the Students Like Learning Mathematics Scale (SLLSS) in the TIMSS 2015 8th grade student survey was determined as 19.6 out of 9-36 points. The overall level of agreement between the cut-off score and the classification of students who like/don’t like learning mathematics using the criterion based on the expression given in the original scale description was found to be 83%. The results concluded that the validity of the standard-setting studies conducted with the artificial neural network method was high. As a result, researchers are recommended to use the Artificial Neural Networks method to determine the cut-off score to be used in the interpretation of the total scores obtained from the achievement test or the total scale scores obtained from the scales.

A novel neural network and sensitivity analysis method for predicting the thermal resistance of heat pipes with nanofluids

It is critical to accurately predict the heat transfer performance of heat pipes with nanofluids for the design of the thermal management system. The current prediction models for the thermal performance of heat pipe with nanofluids are mostly developed based solely on their experimental data, resulting in poor universality and low predictive performance. Moreover, there is a lack of quantitative analysis of the correlation degree of selected input parameters on the thermal performance of heat pipe. The innovation of this work lies in the development of a novel thermal resistance prediction model based on neural network and sensitivity analysis method, characterized by a broad range of inputs and high accuracy. Firstly, a neural network database was established based on our own experimental data and published literature. The effects of the learning rate and the number of nodes in the hidden layer on the prediction performance were investigated. Subsequently, the Garson algorithm was used for sensitivity analysis of the input parameters. The results showed that when the Adam adaptive optimization algorithm and 28 nodes in the hidden layer were used, the neural network model had the highest prediction accuracy, with a mean square error (MSE) of only 0.0020 and a mean absolute error (MAE) of 0.03 K/W, and a mean relative error (MRE) of 7.86 %, demonstrating that the developed model can be used for predicting the performance of heat pipes. The ranking of the influence of each input variable on the thermal resistance from high to low is as follows: heat load, confinement number, inclination angle, cold source temperature, filling ratio, thermal conductivity, and concentration. Moreover, the influence of concentration on the accuracy of the model cannot be ignored. This work provides a direct guidance for developing a performance prediction model for heat pipe with nanofluids with high universality and accuracy.

The role of fixed assets in the modern Russian economy

Subject. The article deals with ensuring the technological sovereignty of Russia. Objectives. The purpose is to determine the impact of characteristics of fixed assets on a number of macroeconomic indicators. Methods. The study employs methods of statistical, neural network, and cluster analysis. Results. The paper established that the growth of fixed assets renewal coefficient has a positive effect on the dynamics of the current total market capitalization of shares of leading Russian companies. Conclusions. The findings can be applied to designing programs for the development of the financial sector of the Russian economy.

Forecasting and Coupled Coordination Analysis of Supply and Demand for Sustainable Talent in Chinese Agriculture

In recent years, China has achieved notable results with its poverty alleviation program, the focus of which is shifting toward the comprehensive promotion of rural revitalization. The role played by sustainable human resources in agriculture is becoming increasingly prominent. In this context, China’s sustainable talent in agriculture is used as the research object, and a neural network analysis method is applied to construct a prediction model of sustainable agricultural talent to forecast its supply and demand. The prediction aims to provide a scientific basis for the strategic planning of talent development for rural revitalization. Based on the forecast results by region and province, we analyzed the level of coordinated development of talent supply and demand to provide a reference for the coordinated development of supply and demand of sustainable talent in agriculture in China. The results showed that a large sustainable agricultural talent demand gap exists in China. The overall talent supply and demand coupling coordination level is low; we found significant differences among different regions and provinces, characterized by decreasing order of the northeast, central, west, and east. According to the socio-economic development level, agricultural economic foundation, and other factors, we divided the provinces into six types for analysis. To promote the coordinated development of sustainable human agricultural resources, talent policy support at the national level is required to reduce the loss of human resources to other countries; at the regional level, the talent environment for rural revitalization should be optimized to increase the attraction of talent. At the provincial level, agricultural and forestry education resources should be created to increase the supply of sustainable agricultural talent.

Prediction method of blast load on underground structure surface based on neural network

The dynamic load in the soil directly leads to the damage of underground structures upon explosions. In this study, a method to predict blast load on underground structure surface based on the neural network was developed to study the load distribution under close-in detonation. First, taking the underground utility tunnel as the experimental structure, 52 groups of field blast tests were conducted on the surface load mechanism, and the surface load data samples were obtained. Second, the key influencing parameters of the reflected blast load were obtained through the dimensional analysis method, and the backpropagation neural network model was constructed based on the test data using the Levenberg–Marquardt algorithm to train and optimize the neural network. Finally, the accuracy of load prediction results was compared and evaluated among the neural network, empirical formula, and nonlinear regression analysis (NRA) methods. It is found that the input parameters of combined variables can further improve the prediction accuracy of the neural network compared with the input parameters of single physical variables. Compared with the empirical formula method and the NRA method, the neural network model with input parameters of combined variables provided the most accurate prediction. The load distribution under typical conditions calculated by the neural network showed that the explosive setting parameters impact the uneven shape of blast load on the structure surface. The increase in explosive equivalent and depth reduces the nonuniformity of load distribution, while the decrease in explosion distance increases the nonuniformity of load distribution.