Prediction Of Behavior Research Articles

Multi-modal information fusion for multi-task end-to-end behavior prediction in autonomous driving

Multi-modal information fusion for multi-task end-to-end behavior prediction in autonomous driving

Dem-driven investigation and AutoML-Enhanced prediction of Macroscopic behavior in cementitious composites with Variable frictional parameters

Dem-driven investigation and AutoML-Enhanced prediction of Macroscopic behavior in cementitious composites with Variable frictional parameters

BiLSTM-based complete stress–strain behavior prediction in monolayer graphene under uniaxial stretching: An integrated molecular dynamics study

BiLSTM-based complete stress–strain behavior prediction in monolayer graphene under uniaxial stretching: An integrated molecular dynamics study

Transforming regression to classification for enhancing extrapolation capacity of data-driven structural behavior prediction

Transforming regression to classification for enhancing extrapolation capacity of data-driven structural behavior prediction

Research on the corrosion behavior prediction model of weathering steel composite bridge stud connectors based on accelerated corrosion and cellular automata

Research on the corrosion behavior prediction model of weathering steel composite bridge stud connectors based on accelerated corrosion and cellular automata

Driving behavior analytics: an intelligent system based on machine learning and data mining techniques

One of the most common causes of road accidents is driver behavior. To reduce abnormal driver behavior, it must be detected early on. Previous research has demonstrated that behavioral and physiological indicators affect drivers' performance. The goal of this study is to consider the feasibility of classifying driver behavior as either aggressive (sudden left or right turns, accelerating and braking), normal (average driving events) or slow (keeping a lower-than-average speed). Innovation in data mining and machine learning (ML) has allowed for the creation of powerful prediction tools. ML techniques have shown potential in predicting driver behavior, with classification being a critical study area. The data set was gathered using the Kaggle platform. This study classifies driver behavior using Orange3 data mining tools and tests several classifiers, including AdaBoost, CN2 rule inducer, and random forest (RF) classifiers. The results showed that AdaBoost was superior in predicting driver behavior, with 100% accuracy, while the classification accuracy in CN2 rule inducer and RF was 99.8% and 95.4%, respectively. These results demonstrate the possibility of early and highly accurate driver behavior prediction and use it to create a ML-based driver behavior detection system.

Attention-enhanced LSTM for high-value customer behavior prediction: Insights from Thailand’s E-commerce sector

Attention-enhanced LSTM for high-value customer behavior prediction: Insights from Thailand’s E-commerce sector

Surrogate modeling for flow simulations using design variable-coded deep learning networks

In recent years, machine learning techniques have emerged as pivotal tools across scientific and engineering disciplines. One notable application is in computational fluid dynamics (CFD), where there is a growing demand for cost-effective alternatives to traditional, resource-intensive simulations. This study explores a surrogate modeling approach known as Design Variable Coded Multilayer Perceptron (DV-MLP), aimed at predicting velocity on unseen geometries using CFD simulation data. The DV-MLP model integrates spatial and design variables directly, offering a mesh-independent, scalable solution for rapid flow prediction across various geometric configurations. This method seeks to replace CFD simulations for geometries resembling those used in training, thereby accelerating the design process. The dataset comprises CFD results from four distinct bump heights. The objective is to train the MLP using data from three bump heights and assess its predictive performance on a middle bump height. The results indicate that the MLP accurately predicts flow features for the middle bump height. This demonstrates its ability to provide precise predictions of near-wall flow behavior for similar geometries, making it a valuable tool for industries requiring swift and reliable CFD predictions for design refinement and optimization. The DV-MLP model is designed primarily for interpolation between geometries and flow conditions represented in the training data, providing accurate predictions for cases that fall within the range of the sampled design space.

Deformation behavior prediction and structural optimization of vertical hydrostatic guideway ram in vertical lathe

Deformation behavior prediction and structural optimization of vertical hydrostatic guideway ram in vertical lathe

Machine Learning Assisted Corrosion Behaviour Prediction of Dual-Engineered Ti6Al4V Alloy

Machine Learning Assisted Corrosion Behaviour Prediction of Dual-Engineered Ti6Al4V Alloy

Deep Learning-Based Framework for Social Media User Behavior Analysis and Prediction

The interconnection of billions of users through social media platforms has made them indispensable in contemporary communication. It is essential to analyze and forecast user behavior on these platforms to improve user experience, implement targeted advertising, and reduce unpleasant interactions. This study aimed to create deep learning-based algorithms for social media user behavior analysis and prediction. The first data corpus we analyzed comprises tweets on the general elections in India in 2019. The data prepossessing process used Z-score normalization and missing value-relevant features to normalize the data’s scale. Features were extracted from high-dimensional data using the linear regression (LR) model, which preserves important information representations for the prediction model to improve computing efficiency and reduce overfitting. The term frequency-inverse document frequency (TF-IDF) numerical statistic was used to compute this. We suggested optimizing the socially restricted Boltzmann machine dove particle swarm with bi-directional long short-term memory (SRBM-DPSO-bi-LSTM) to analyze and predict social media user behavior. Training data optimize the specified deep learning algorithms for essential performance measures and frequently enhance the prediction system to keep up with social media user behavior. This will make it possible to evaluate how well the suggested solution works regarding prediction rate, accuracy, recall, F1-score, and the AUC-ROC curve. As a result, this proposed deep learning approach to prediction data to improve Social Media User Behavior.

Research Progress and Technology Outlook of Deep Learning in Seepage Field Prediction During Oil and Gas Field Development

As the development of oilfields in China enters its middle-to-late stage, the old oilfields still occupy a dominant position in the production structure. The seepage process of reservoirs in the high Water Content Period (WCP) presents significant nonlinear and non-homogeneous evolution characteristics, and the traditional seepage-modeling methods are facing the double challenges of accuracy and adaptability when dealing with complex dynamic scenarios. In recent years, Deep Learning technology has gradually become an important tool for reservoir seepage field prediction by virtue of its powerful feature extraction and nonlinear modeling capabilities. This paper systematically reviews the development history of seepage field prediction methods and focuses on the typical models and application paths of Deep Learning in this field, including FeedForward Neural networks, Convolutional Neural Networks, temporal networks, Graphical Neural Networks, and Physical Information Neural Networks (PINNs). Key processes based on Deep Learning, such as feature engineering, network structure design, and physical constraint integration mechanisms, are further explored. Based on the summary of the existing results, this paper proposes future development directions including real-time prediction and closed-loop optimization, multi-source data fusion, physical consistency modeling and interpretability enhancement, model migration, and online updating capability. The research aims to provide theoretical support and technical reference for the intelligent development of old oilfields, the construction of digital twin reservoirs, and the prediction of seepage behavior in complex reservoirs.

Prediction of Self-Care Behaviors in Patients Using High-Density Surface Electromyography Signals and an Improved Whale Optimization Algorithm-Based LSTM Model

Prediction of Self-Care Behaviors in Patients Using High-Density Surface Electromyography Signals and an Improved Whale Optimization Algorithm-Based LSTM Model

Evacuee Behavior Modeling during Robot-Guided Evacuations

Abstract Evacuation robots hold promise for facilitating efficient and safe building evacuations during emergencies. While studies have demonstrated that people will follow an evacuation robot [Nayyar 2019 28th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN):1–6, 1, Robinette et al. 2016 11th ACM/IEEE International Conference On:101–108, 2], many practical hurdles remain. This paper uses a human subject study in a physical environment to investigate robot-guided evacuation of individuals versus small groups and considers different strategies for using robots to guide evacuees to exits. We further show that the data collected from these human subject studies can be used to train evacuee motion models which accurately predict the movement of the evacuee (9.9 cm mean error) while following the robot. We further show that this model can be used to predict the motion of an evacuee in a different environment and show that the accuracy of our model is superior to the more standard social forces model (SFM). The results from this research will contribute to the creation of evacuation robots and to the modeling and prediction of evacuee behavior.

Ethical issues in the use of genetic predictions of aggressive behavior in the criminal justice system: a systematic review.

The use of genetic predictions of aggressive behavior in the criminal justice system remains a subject of ongoing debate. Since behavioral genetic evidence is often used in criminal defense arguments, it is crucial to critically examine the ethical challenges associated with its application. This article seeks to identify and analyze these ethical concerns to ensure the responsible and equitable integration of genetic testing, when deemed necessary, into the judiciary system. A systematic review was conducted using PubMed, Web of Science, and Scopus, supplemented by manual searches of reference lists to identify additional relevant studies. The search yielded 1,023 publications, 12 of which met the inclusion criteria. Seven key ethical concerns were identified: the risks of discrimination, stigmatization, eugenic reasoning, deterministic interpretations, overestimation of dangerousness, privacy violations, and medicalization, along with the risks posed by limited scientific literacy among legal professionals. The ethical challenges associated with genetic predictions of aggressive behavior underscore the need for a critical and multidisciplinary approach to their use in the criminal justice system. Collaboration among bioethicists, legal scholars, scientists, and communication experts is crucial to prevent misuse and reduce potential biases. Such an approach will help ensure that genetic insights are ethically applied, accurately interpreted, and used to promote justice rather than exacerbate systemic inequalities.

Experimental Studies to Evaluate the Effects of Different Unloading Stress Paths on Strength Properties of Unconsolidated Sands

The mechanical behavior of rocks under loading conditions depends on stress path and magnitude. With increasing load, rocks have an elasto-plastic behavior. Within the loading yield surface, constitutive models assume that rocks behave elastically and are independent of the stress path and magnitude (e.g., Mohr–Coulomb models). We performed tests on unconsolidated sands (no cementation), and under both loading and unloading conditions. We mapped the loading yield surface using a multi-stage triaxial test with the yield criterion as the point of positive dilatancy. We studied the yield behavior of the two different unloading stress paths: a constant axial stress unloading test (reducing mean effective stress and increasing shear stress) and a constant shear stress unloading test (reducing mean effective stress and keeping shear stress constant). The results show that unloading-based tests reach yield point at a lower shear stress than expected from the loading-based yield surface. The unloading-based yield surface is also dependent on the stress path. The application of this research includes a prediction of the geomechanical behavior of unconsolidated sands under injection conditions. Often, a constitutive model derived from loading stress paths is used for injection with the ad hoc assumption that the loading and unloading models are identical. These constitutive models provide results for injector design parameters, injection performance prediction, and safe injection envelopes. Therefore, it is essential to have accurate constitutive models that are representative of unloading stress paths. In calibrating these models, we demonstrated that the yield criterion (point of positive dilatancy) is reached before the loading-based yield surface during injection (decrease in mean effective stress) is reached. We also developed a minimum yield surface model. With a calibration using three tests, this model can predict the yield point for any stress path and at any initial stress state (within the bounds of the experiments).

The Bidirectional Relationship Between Sleep Quality and Aggressive Behavior: Within-Person Mediated Effect of Self-Control.

The correlation between sleep quality, self-control, and aggressive behavior has been assessed in previous empirical studies. However, the interrelationship and underlying mechanisms of sleep alterations and aggressive behavior among adolescents at the within-person level, particularly within the context of Chinese culture, have rarely been investigated. Using a random intercept cross-lagged panel model (RI-CLPM), this longitudinal study aims to (i) distinguish between-person and within-person effects when examining the interplay between sleep quality and aggressive behavior among adolescents and (ii) assess the potential mediating role of self-control in this relationship. The present sample consisted of 1240 Chinese adolescents, including 614 girls, with a mean age of 12.72 years (SD = 0.68) at baseline. Data were collected across four waves over 2 years, with an interval of 6 months between consecutive time points. The RI-CLPM results indicated that sleep quality directly predicted aggressive behavior at the within-person level and vice versa, demonstrating a bidirectional predictive relationship between the two constructs. Moreover, self-control mediated the prediction of aggressive behavior by sleep quality, but not the reverse. These findings underscore a reciprocal influence between sleep quality and aggressive behavior among Chinese youth, confirming that low levels of self-control significantly mediate the effect of sleep quality on aggression.

Detecting Workplace Hubris: A Machine Learning Approach to Narcissism Identification. The Case of the Healthcare Industry in the Emerging Markets

Despite the extensive research on narcissism and its origin, the world of health practice, risk factors, as well as the case in a developing country like Morocco, is a new untapped area. This work explores uncharted territory as it attempts to replace the existing social behavior prediction tools with different machine learning models that promise the best approach to narcissist behavior prediction by identifying psychological features characteristic of narcissist personalities. Among different machine learning models used in this study, Support Vector Machine (SVM) shows the highest metrics with an accuracy of 0.910, precision of 0.890, and recall of 0.880. SVM reveals that vanity, self-sufficiency, authority, and exhibitionism are the best predictors of narcissism in organizational settings.

Analytical model of Johnson–Kendall–Roberts adhesive contact with coating roughness on the release of test mass in TianQin

Abstract The surface of the coating or substrate structure has a rough topological structure over multiple scales, which significantly affects the actual contact area and adhesion strength. In order to investigate potential interferences during the test mass (TM) release process of the locking and release mechanism in TianQin space-borne gravitational wave detection project, an analytical model of the Johnson–Kendall–Roberts (JKR) adhesive contact with coating roughness between the TM and the release tip has been developed. This work presents the integration of the JKR theory with the Greenwood–Williamson rough surface contact theory to address adhesion challenges and influence of roughness in space-borne gravitational wave detection systems. The comprehensive model can be used to analyze how various factors such as surface shape, material properties, coating thickness and roughness affect adhesion. By introducing a numerical approach based on the Newton down-hill method to solve coupled equations for gap, stress, and load balance, the model achieves high-precision predictions of adhesion behavior under multi-scale roughness effects. By establishing relationships between force, contact half-width, and penetration depth, it provides a theoretical basis for optimizing the TM release process and the design of the locking and release mechanism. Future work will validate predictions via high-precision torsion pendulum experiments.

Impact of Gender and Demographic Factors on Smoking Behavior Prediction

The relationship between smoking habits and gender, that is affected by demographic variables like age, income level, and education level, is illustrated. Using the Chi-square test for independence, we identified significant associations with gender on smoking behavior. EDA further revealed crucial trends and disparities in prevalence of smoking among different groups of demographic variables. We develop a classification model to identify the significant predictors for smoking behavior from the factors mentioned above. Further, we conduct Principal Component Analysis for reducing the dimension of data to see which factors are more significantly influential for smoking behavior. The results of the research give us the understanding that can be applied towards the formulation of targeted public health strategies that could effectively reduce the prevalence of smoking.