Neural Network Analysis Research Articles

MODELLING ROUNDABOUT ENTRY CAPACITY FOR MIXED TRAFFIC FLOW USING ANN: A CASE STUDY IN INDIA

Roundabouts, as an unsignalized intersection, have an effective preventative measure designed to control straight-line crashes. Efficient traffic flow in cities depends upon appropriate capacity estimation of roundabouts. This study attempts to develop models for roundabout entry capacity by applying Artificial Neural Network (ANN) analysis for mixed traffic flow conditions. Data was gathered from 27 roundabouts spread across India. The influence area for gap acceptance (INAGA) concept was used as a graphical method to identify critical gap (Tc) of entry flow at roundabouts. This study indicated that the Bayesian Regularisation Neural Network (BRNN) based model has the best R2 and RMSE of 0.97 and 167.8. The connection weight approach and Garson algorithm evaluate the significance of each explanatory variable and identify follow-up time (Tf) as a critical parameter with values of 11.10 and 21.15%, respectively.

Artificial neural network modelling and experimental investigations of malachite green adsorption on novel carboxymethyl cellulose/β-cyclodextrin/nickel cobaltite composite

This study presents a novel polymer nanocomposite based on carboxymethyl cellulose and β-cyclodextrin crosslinked with succinic acid (CMC-SA-β-CD) containing nickel cobaltite (NCO) nano-reinforcement. Various analytical techniques have been employed to investigate the structural, thermal, and morphological features of the resulting nanocomposite. The CMC-SA-β-CD/NCO nanocomposite has been utilized as an adsorbent for the removal of bisphenol-A (BPA, R% <40 %), malachite green (MG, R% > 75 %)), and Congo red (CR, no adsorption) from the synthetic wastewater. The study systematically explored the impact of various parameters on the adsorption process, and the interactions between MG and CMC-SA-β-CD/NCO were discussed. The adsorption data were fitted to different models to elucidate the kinetics and thermodynamics of the adsorption process. An artificial neural network (ANN) analysis was employed to train the experimental dataset for predicting adsorption outcomes. Despite a low BET surface area (0.798 m2 g−1), CMC-SA-β-CD/NCO was found to exhibit high MG adsorption capacity. CMC-SA-β-CD/NCO exhibited better MG adsorption performance at pH 5.5, 40 mg L−1 MG dye concentration, 170 min equilibrium time, 20 mg CMC-SA-β-CD/NCO dose with more than 90 % removal efficiency. Moreover, the thermodynamic studies suggest that the adsorption of MG was exothermic with ΔH° value −9.93 ± 0.76 kJ mol−1. The isotherm studies revealed that the Langmuir model was the best model to describe the adsorption of MG on CMC-SA-β-CD/NCO indicating monolayer surface coverage with Langmuir adsorption capacity of 182 ± 4 mg g−1. The energy of adsorption (11.4 ± 0.8 kJ mol−1) indicated chemisorption of MG on the composite surface. The kinetics studies revealed that the pseudo-first-order model best described the adsorption kinetics with qe = 86.7 ± 2.9 mg g−1. A good removal efficiency (>70 %) was retained after five regeneration reuse cycles. The ANN-trained data showed good linearity between predicted and actual data for the adsorption capacity (R-value>0.99), indicating the reliability of the prediction model. The developed nanocomposite, composed predominantly of biodegradable material, is facile to synthesize and exhibited excellent monolayer adsorption of MG providing a new sustainable adsorbent for selective MG removal.

Enhancing Sales Prediction for MSMEs: A Comparative Analysis of Neural Network and Linear Regression Algorithms

Enhancing Sales Prediction for MSMEs: A Comparative Analysis of Neural Network and Linear Regression Algorithms

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.

Comparative analysis of convolutional neural network and DenseNet121 transfer learning in agriculture focusing on crop leaf disease identification

PurposePlant cultivation holds a pivotal role in agriculture, necessitating precise disease identification for the overall health of plants. This research conducts a comprehensive comparative analysis between two prominent deep learning algorithms, convolutional neural network (CNN) and DenseNet121, with the goal of enhancing disease identification in tomato plant leaves.Design/methodology/approachThe dataset employed in this investigation is a fusion of primary data and publicly available data, covering 13 distinct disease labels and a total of 18,815 images for model training. The data pre-processing workflow prioritized activities such as normalizing pixel dimensions, implementing data augmentation and achieving dataset balance, which were subsequently followed by the modeling and testing phases.FindingsExperimental findings elucidated the superior performance of the DenseNet121 model over the CNN model in disease classification on tomato leaves. The DenseNet121 model attained a training accuracy of 98.27%, a validation accuracy of 87.47% and average recall, precision and F1-score metrics of 87, 88 and 87%, respectively. The ultimate aim was to implement the optimal classifier for a mobile application, namely Tanamin.id, and, therefore, DenseNet121 was the preferred choice.Originality/valueThe integration of private and public data significantly contributes to determining the optimal method. The CNN method achieves a training accuracy of 90.41% and a validation accuracy of 83.33%, whereas the DenseNet121 method excels with a training accuracy of 98.27% and a validation accuracy of 87.47%. The DenseNet121 architecture, comprising 121 layers, a global average pooling (GAP) layer and a dropout layer, showcases its effectiveness. Leveraging categorical_crossentropy as the loss function and utilizing the stochastic gradien descent (SGD) Optimizer with a learning rate of 0.001 guides the course of the training process. The experimental results unequivocally demonstrate the superior performance of DenseNet121 over CNN.

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.

Demonstration of angioplasty using a balloon catheter in tetra-hybrid nano-bloodstream within an electrified stenotic arterial cavity under a magnetic field: Artificial neural network analysis

This article focuses on the critical role of advanced computational techniques in enhancing medical diagnostics and treatment planning, especially in procedures like angioplasty. It presents a detailed examination of ballooned catheterization (balloon angioplasty) in a bloodstream carrying tetra-hybrid nanoparticles within a negatively charged stenotic arterial cavity, influenced by a magnetic field. The study models the streaming of blood, considering factors like heat generation, Joule heating, interfacial nanolayers, nanoparticle size, and velocity and thermal slip conditions. The model simplifies complex dynamics through the lubricant and Debye–Hückel linearization approaches and employs the homotopy perturbation method (HPM) for solving dimensionless equations. Mathematica and Matlab are used to illustrate key parameters like velocity, temperature, pressure gradient, wall shear stress (WSS), heat transfer coefficient (HTC), and blood bolus trapping. The study reveals that increasing the balloon catheter’s height boosts blood velocity in the annulus’s central zone, while a thicker interfacial nanolayer diminishes blood temperature and HTC. Larger nanoparticles elevate the blood temperature profile, and increased electro-osmotic parameters enhance the pressure gradient. The tetra-hybrid nano-blood (THNB) shows superior HTC compared to other forms. The study also delves into bolus trapping phenomena, emphasizing the significant changes induced by electro-osmotic forces. Additionally, an artificial neural network (ANN) model, developed using HPM-derived data, accurately predicts WSS and HTC, achieving high accuracy rates of 99.97% in testing and 99.92% in validation for THNB flow. This study paves the way for more targeted, safer, and efficient use of balloon catheters in treating arterial blockages, significantly impacting cardiovascular medicine and patient care.

The Performance Analysis of Graph Neural Network (GNN) and Convolutional Neural Network (CNN) Algorithms for Cyberbullying Detection in Twitter Comments

Cyberbullying incidents have surged due to the expansion of social media network and advancements in internet technology, presenting a substantial challenge in online communities. Previous studies employing Support Vector Machine (SVM) techniques have exhibited promising outcomes, achieving a superior accuracy of 71.25%. However, recognizing the dynamic nature of cyberbullying behaviors and the necessity for more robust detection methodologies, this research explores cyberbullying detection on Twitter utilizing Convolutional Neural Network (CNN) and Graph Neural Network (GNN). The selection of CNN and GNN is motivated by the deficiencies observed in prior SVM-based approaches and the capacity of neural network to capture intricate patterns in textual and network data. The GNN consistently outperforms CNN in terms of F1 score, accuracy, precision, and recall. With only 20 epochs, GNN achieves an accuracy of 80.25%, surpassing CNN's 68.43%. Through GNN optimization, its accuracy reaches 89.04% after 100 epochs, underscoring its efficacy in Twitter cyberbullying detection.

Analysis of convolutional neural networks for fronto-temporal dementia biomarker discovery.

Frontotemporal lobe dementia (FTD) results from the degeneration of the frontal and temporal lobes. It can manifest in several different ways, leading to the definition of variants characterised by their distinctive symptomatologies. As these variants are detected based on their symptoms, it can be unclear if they represent different types of FTD or different symptomatological axes. The goal of this paper is to investigate this question with a constrained cohort of FTD patients in order to see if the heterogeneity within this cohort can be inferred from medical images rather than symptom severity measurements. An ensemble of convolutional neural networks (CNNs) is used to classify diffusion tensor images collected from two databases consisting of 72 patients with behavioural variant FTD and 120 healthy controls. FTD biomarkers were found using voxel-based analysis on the sensitivities of these CNNs. Sparse principal components analysis (sPCA) is then applied on the sensitivities arising from the patient cohort in order to identify the axes along which the patients express these biomarkers. Finally, this is correlated with their symptom severity measurements in order to interpret the clinical presentation of each axis. The CNNs result in sensitivities and specificities between 83 and 92%. As expected, our analysis determines that all the robust biomarkers arise from the frontal and temporal lobes. sPCA identified four axes in terms of biomarker expression which are correlated with symptom severity measurements. Our analysis confirms that behavioural variant FTD is not a singular type or spectrum of FTD, but rather that it has multiple symptomatological axes that relate to distinct regions of the frontal and temporal lobes. This analysis suggests that medical images can be used to understand the heterogeneity of FTD patients and the underlying anatomical changes that lead to their different clinical presentations.

Deep Learning Algorithms for Longitudinal Driving Behavior Prediction: A Comparative Analysis of Convolutional Neural Network and Long–Short-Term Memory Models

Deep Learning Algorithms for Longitudinal Driving Behavior Prediction: A Comparative Analysis of Convolutional Neural Network and Long–Short-Term Memory Models

Towards a Comprehensive Framework for Regional Transportation Land Demand Forecasting: Empirical Study from Yangtze River Economic Belt, China

China is currently experiencing rapid expansion in its transportation land. To promote sustainable land use, accurately estimating transportation land demand is crucial. This study aims to develop a comprehensive framework for urban transportation land forecasting within the Yangtze River Economic Belt (YREB), providing support for optimizing regional land allocation. Employing methods such as meta-analysis, statistical analysis, and BP neural network analysis, this study forecasts the transportation land demand of 127 cities in the YREB. The study findings indicate that cities with high transportation land demand are mainly distributed in the middle and upper reaches of the Yangtze River. Moreover, the growth rate of transportation land in the upper reaches significantly outstrips that in the middle and lower reaches, suggesting a focus shift in transportation infrastructure construction toward the upper regions. Additionally, some cities within the YREB face a mismatch between the supply and demand of transportation land, necessitating proactive adjustments to their land supply plans to achieve a balance between supply and demand. The main contribution of this study is the development of a comprehensive and adaptable framework that guides the development of future strategies for optimal land allocation by forecasting transportation land demand at a regional level.

Design and evaluation of piperidine carboxamide derivatives as potent ALK inhibitors through 3D-QSAR modeling, artificial neural network and computational analysis

Tumor stands as one of the principal contributors to global mortality. As research into tumor treatments advances, tumor inhibitors emerge as pivotal milestones in tumor therapy. Among these inhibitors, Anaplastic Lymphoma Kinase (ALK), a receptor tyrosine kinase, is critical owing to its close association with tumor cell proliferation and growth, which renders it a critical therapeutic target. This work systematically explores the relationship between the chemical structures of 36 piperidine carboxamide derivatives and their efficacy in inhibiting Karpas-299 tumor cell activity by employing a rigorous 3D-QSAR modeling approach. A robust Topomer CoMFA model was generated and was meticulously validated through ANN neural network analysis (q2 = 0.597, r2 = 0.939, F = 84.401, N = 4, SEE = 0.268). Based on the model, 60 new compounds with desirable inhibitory activities were successfully designed. Combined with Lipinski’s rule and ADMET criteria alongside molecular docking and dynamics simulations, a lead compound with high inhibitory activity and good drug-likeness was selected. Further computational analyses, encompassing free energy landscape and binding free energy calculations, provided compelling evidence of the stable binding conformation of the lead compound and the superior affinity with the target protein at the active site, underscoring its potential therapeutic utility. In summary, this investigation offers valuable insights and methodological guidance for advancing tumor therapy and underscores the promise of piperidine carboxamide derivatives as prospective ALK inhibitors.

Efficient construction and convergence analysis of sparse convolutional neural networks

In this paper, a new variant of convolutional neural network (CNN) based on L1 regularization is proposed. The main consideration is to generate a sparse weight matrix, i.e., to generate a sparse neural network model. To some extent, L1 regularization prevents the occurrence of overfitting and effectively improves the learning efficiency of the network. Unfortunately, the L1 regularization term is non-smooth, which leads to the occurrence of oscillations in experiments and poses a great challenge to the theoretical analysis. So, we overcome this drawback by using polishing techniques and rigorously prove the monotonicity of the error function and the strong and weak convergence theorems of the algorithm. Finally, numerical simulations on several data sets support our theoretical results and the superiority of the proposed algorithm.

Synchronization analysis of fuzzy inertial neural networks with time‐varying delays via non‐reduced order method

SummaryIn this article, the asymptotic synchronization of a class of fuzzy inertial neural networks (FINNs) with time‐varying delays is investigated. First, the direct analysis approach is applied to replace the accustomed variable transformation and the reduced‐order method for addressing the inertial term. Second, a suitable Lyapunov function and control scheme are devised to obtain several new sufficient conditions to guarantee the asymptotic synchronization of the class of FINNs with time‐varying delays. It turns out that the obtained criteria are simpler and more effective. Meanwhile, some numerical examples demonstrate the effectiveness of the proposed strategies and verify the theoretical results.

Method for Constructing Neural Network Means for Recognizing Scenes of Political Extremism in Graphic Materials of Online Social Networks

Countering the spread of calls for political extremism through graphic content on online social networks is becoming an increasingly pressing problem that requires the development of new technological solutions, since traditional approaches to countering are based on the results of recognizing destructive content only in text messages. Since in modern conditions neural network tools for analyzing graphic information are considered the most effective, it is assumed that it is advisable to use such tools for analyzing images and video materials in online social networks, taking into account the need to adapt them to the expected conditions of use, which are determined by the wide variability in the size of graphic content, the presence of typical interference, limited computing resources of recognition tools. Using this thesis, a method has been proposed that makes it possible to implement the construction of neural network recognition tools adapted to the specified conditions. For recognition, the author's neural network model was used, which, due to the reasonable determination of the architectural parameters of the low-resource convolutional neural network of the MobileNetV2 type and the recurrent neural network of the LSTM type, which makes up its structure, ensures high accuracy of recognition of scenes of political extremism both in static images and in video materials under limited computing conditions resources. A mechanism was used to adapt the input field of the neural network model to the variability of the size of graphic resources, which provides for scaling within acceptable limits of the input graphic resource and, if necessary, filling the input field with zeros. Levelling out typical noise is ensured by using advanced solutions in the method for correcting brightness, contrast and eliminating blur of local areas in images of online social networks. Neural network tools developed on the basis of the proposed method for recognizing scenes of political extremism in graphic materials of online social networks demonstrate recognition accuracy at the level of the most well-known neural network models, while ensuring a reduction in resource intensity by more than 10 times. This allows the use of less powerful equipment, increases the speed of content analysis, and also opens up prospects for the development of easily scalable recognition tools, which ultimately ensures an increase in security and a reduction in the spread of extremist content on online social networks. It is advisable to correlate the paths for further research with the introduction of the Attention mechanism into the neural network model used in the method, which will make it possible to increase the efficiency of neural network analysis of video materials.

A Comparative Analysis of Polynomial Regression and Artificial Neural Networks for Prediction of Lighting Consumption

A Comparative Analysis of Polynomial Regression and Artificial Neural Networks for Prediction of Lighting Consumption

Research on a Data Preprocessing Method for a Vehicle-Mounted Solar Occultation Flux–Fourier Transform Infrared Spectrometer

A vehicle-mounted solar occultation flux–Fourier transform infrared spectrometer uses the sun as an infrared light source to quantify molecular absorption in the atmosphere. It can be used for the rapid three-dimensional monitoring of pollutant emissions and the column concentration monitoring of greenhouse gases. The system has the advantages of high mobility and a capacity for noncontact measurement and measurement over long distances. However, in vehicle-mounted applications, vehicle bumps and obstacles introduce aberrations in the measured spectra, affecting the accuracy of gas concentration inversion results and flux calculations. In this paper, we propose a spectral data preprocessing method that combines a self-organizing mapping neural network and correlation analysis to reject anomalous spectral data measured by the solar occultation flux–Fourier transform infrared spectrometer during mobile observations. Compared to the traditional method, this method does not need to adjust the comparison threshold and obtain the training spectra in advance and has the advantage of automatically updating the weights without the need to set fixed correlation comparison coefficients. The accurate identification of all anomalous simulated spectra in the simulation experiments proved the effectiveness of the method. In the vehicle-mounted application experiment, 342 anomalous spectra were successfully screened from 1739 spectral data points. The experimental results show that the method can improve the accuracy of gas concentration measurement results and can be applied to a vehicle-mounted solar occultation flux–Fourier transform infrared spectrometer system to meet the preprocessing needs of a high number of spectral data in mobile monitoring.

Fitness Landscape Analysis of Product Unit Neural Networks

A fitness landscape analysis of the loss surfaces produced by product unit neural networks is performed in order to gain a better understanding of the impact of product units on the characteristics of the loss surfaces. The loss surface characteristics of product unit neural networks are then compared to the characteristics of loss surfaces produced by neural networks that make use of summation units. The failure of certain optimization algorithms in training product neural networks is explained through trends observed between loss surface characteristics and optimization algorithm performance. The paper shows that the loss surfaces of product unit neural networks have extremely large gradients with many deep ravines and valleys, which explains why gradient-based optimization algorithms fail at training these neural networks.

Research on anomaly detection in oil chromatography based on convolutional neural network and grey relational analysis algorithm

Abstract The power transformer is a crucial voltage conversion node and a core device in substations. Currently, dissolved gas analysis (DGA) in oil is a practical and validated method for assessing the operational status of transformers. In this study, addressing issues such as insufficient data mining from single dissolved gas measurements, we employed a combination of Convolutional Neural Network and Grey Relational Analysis algorithm to enhance the reliability of identifying abnormal dissolved gas data in transformer oil. This approach improves the sensitivity and accuracy of transformer fault detection. The study includes data processing and result analysis, indicating that the proposed algorithm effectively identifies abnormal dissolved gas data in transformer oil.

Possibilities of using neural network analysis in the diagnosis of dry eye syndrome

The prevalence rate of dry eye syndrome varies from 6.5 to 95 %. Diagnostic criteria are based on different methods and/or their combinations and are characterized by heterogeneity.The aim of the study. To identify the risk factors for the development of dry eye syndrome in order to create a technology for early diagnosis of the degree of the disease in young people without concomitant ocular and general somatic pathology.Materials and methods. Fifty patients aged 24 [22; 27] years were examined. We carried out an ophthalmological examination, including autorefractometry, visometry, biomicroscopy, the Norn test, a survey using the author’s questionnaire, and an assessment of the degree of dry eye syndrome using the Ocular Surface Disease Index (OSDI). Three study groups were formed: control group (OSDI = 0–13 points); group 1 – patients with OSDI = 14–22 points; group 2 – patients with OSDI &gt; 22 points.Results. When examining presented independent variables, screen time had the highest normalized importance (100 %), followed by tear film breakup time (58.4 %), smoking (24.3 %), night shifts (22.5 %) and using soft contact lenses (11.1 %). The technology for early diagnosis of the degree of dry eye syndrome is implemented on the basis of a multilayer perceptron, the percentage of incorrect predictions during its training process was 8.0 %. The structure of the trained neural network included 8 input neurons (the value of screen time and tear film breakup time, the presence or absence of smoking, night shifts and/or the use of soft contact lenses), two hidden layers containing 3 and 2 units, respectively, and 3 output neurons.Conclusion. The proposed neural network has no difficulties in assessing the early diagnosis of the severity of dry eye syndrome and can be used in clinical practice.