TARGE: large language model-powered explainable hate speech detection by Muhammad Haseeb Hashir, Memoona, Sung Won Kim

“…We introduce a novel framework wherein large language models (LLMs) generate explicit rationales by identifying and analyzing critical textual features indicative of hate speech. …”
Get full text

Modeling the Internet of Things network for monitoring audio information on the Amazon platform by U. A. Vishniakou, B. H. Shaya

“…The algorithm for connecting devices to the AWS platform is given: creating a device certificate on the platform, creating a security policy, rules for processing information received from devices, and testing the network. The features of the algorithm for modeling the readings of sound information sensors on a smartphone are shown, steps are given for organizing its communication with the platform, performing security procedures, sending data in the form of an MQTT message, and displaying the captured audio information.…”
Get full text

Unraveling disulfidptosis for prognostic modeling and personalized treatment strategies in lung adenocarcinoma by Xiangyu Xu, Bingbing Zhang, Jin Zhang, Hongbiao Ma

“…The nomogram, incorporating the risk model with clinical features, provided a reliable tool for predicting one-year (AUC 0.77), three-year (AUC 0.75), and five-year (AUC 0.78) survival rates. …”
Get full text

DualAD: Dual adversarial network for image anomaly detection⋆ by Yonghao Wan, Aimin Feng

“…Additionally, the authors employ dual adversarial learning to model the distribution of normal data. On the one hand, adversarial learning was implemented during the reconstruction process to obtain higher‐quality reconstruction samples, thereby preventing the effects of blurred image reconstructions on model performance. …”
Get full text

Accurate and Efficient Numerical Simulation of Land Models Using SUMMA With SUNDIALS by Raymond J. Spiteri, Ashley E. VanBeusekom, Kyle Klenk, Reza Zolfaghari, Sean J. Trim, Wouter J. M. Knoben, Andrew M. Ireson, Martyn P. Clark

“…Being able to efficiently perform more reliable simulations makes the SUMMA‐SUNDIALS model a powerful tool for improving our understanding of the terrestrial component of the Earth System.…”
Get full text

A Multitask Deep Learning Model for Predicting Myocardial Infarction Complications by Fazliddin Makhmudov, Normakhmad Ravshanov, Dilshot Akhmedov, Oleg Pekos, Dilmurod Turimov, Young-Im Cho

“…The model was trained on a dataset of 1700 patients, encompassing 111 clinical and demographic features. …”
Get full text

Probabilistic seismic hazard mapping for Bangladesh using updated source models by Anika Samm-A, A. S. M. Maksud Kamal, Md. Shakhawat Hossain, Md. Zillur Rahman

“…Despite identifying several tectonic features, their detailed characterization (e.g. slip type, slip rate, movement direction) remains unclear, leading to a lack of well-defined seismic source models and Ground Motion Prediction Equations (GMPEs) specific to the region to assess seismic hazards. …”
Get full text

Modeling Local Demand for Mobile Spectrum: An Interpretable Machine Learning Approach by Janaki Parekh, Elizabeth Yackoboski, Amir Ghasemi, Halim Yanikomeroglu

“…The top-performing model successfully achieves an R2 of 0.76 and a Root Mean Square Error of 51.02 on the hold-out test set. …”
Get full text

Explainable AI for Enhancing Efficiency of DL-Based Channel Estimation by Abdul Karim Gizzini, Yahia Medjahdi, Ali J. Ghandour, Laurent Clavier

“…Hence the designed XAI-CHEST delivers a smart low-complex one-shot input feature selection methodology for high-dimensional model input that can further improve the overall performance while optimizing the architecture of the employed model. …”
Get full text

A New Hybrid Intelligent Method for Accurate Short Term Electric Power Production Forecasting From Uncertain Renewable Energy Resources by Gholamreza Memarzadeh, Farshid Keynia, Faezeh Amirteimoury, Rasoul Memarzadeh, Hossein Noori

“…The suggested hybrid model employs the Modified Relief-Mutual Information (MRMI) feature selection technique to identify the most influential input data for prediction. …”
Get full text

Multimedia Pop Music Teaching Model Integrating Semifinished Teaching Strategies by Kangtan Dong

Get full text

Machine Learning Tool for Analyzing Finite Buffer Queueing Systems by Attahiru Sule Alfa, Haitham Abu Ghazaleh

“…These network systems and their performance measures are typically analyzed using queueing-based models. …”
Get full text

Optimal Activity Recognition Framework Based on Improvement of Regularized Neighborhood Component Analysis (RNCA) by Norazman Shahar, Muhammad Amir As'Ari, Tan Tian Swee, Nurul Fathia Ghazali

“…In essence, the combined strength of RNCA and MRMR provides a versatile and effective approach for extracting meaningful features and enhancing the overall performance of machine learning models.…”
Get full text

Modeling the heating of the reservoir near-wellbore zone using an induction heater by Denis V. Kosmylin, Vladimir Ya. Fedotov, Filyus F. Davletshin, Ramil F. Sharafutdinov, Rim A. Valiullin

“…It was found that the reservoir heating with a value of more than 5 degrees Celsius is achieved in the near-wellbore zone of the reservoir with a radius of 0.2–0.3 m, depending on the thermal conductivity of the rocks. The authors performed the experimental studies of a thermal field in a physical model of a well with an induction heater. …”
Get full text

Investigation into the Hyperparameters of Error-Based Adaptive Sampling Approach for Surrogate Modeling by Leonid Legashev, Sergey Tolmachev, Irina Bolodurina, Alexander Shukhman, Lyubov Grishina

“…The maximum dry bulb temperature on a winter and summer day and the wind speed on a winter day were the most important features of the built surrogate model with 492, 483 and 443 gain values of the feature importance method, respectively.…”
Get full text

Video temporal perception characteristics based just noticeable difference model by Yafen1 XING, Haibing YIN, Hongkui WANG, Qionghua LUO

“…The existing temporal domain JND(just noticeable distortion) models are not sufficient to depict the interaction between temporal parameters and HVS characteristics, leading to insufficient accuracy of the spatial-temporal JND model.To solve this problem, feature parameters that can accurately describe the temporal characteristics of the video were explored and extracted, as well as a homogenization method for fusing heterogeneous feature parameters, and the temporal domain JND model based on this was improved.The feature parameters were investigated including foreground and background motion, temporal duration along the motion trajectory, residual fluctuation intensity along motion trajectory and adjacent inter-frame prediction residual, etc., which were used to characterize the temporal characteristics.Probability density functions for these feature parameters in the perception sense according to the HVS(human visual system) characteristics were proposed, and uniformly mapping the heterogeneous feature parameters to the scales of self-information and information entropy to achieve a homogeneous fusion measurement.The coupling method of visual attention and masking was explored from the perspective of energy distribution, and the temporal-domain JND weight model was constructed accordingly.On the basis of the spatial JND threshold, the temporal domain weights was integrated to develop a more accurate spatial-temporal JND model.In order to evaluate the performance of the spatiotemporal JND model, a subjective quality evaluation experiment was conducted.Experimental results justify the effectiveness of the proposed model.…”
Get full text

Video temporal perception characteristics based just noticeable difference model by Yafen1 XING, Haibing YIN, Hongkui WANG, Qionghua LUO

“…The existing temporal domain JND(just noticeable distortion) models are not sufficient to depict the interaction between temporal parameters and HVS characteristics, leading to insufficient accuracy of the spatial-temporal JND model.To solve this problem, feature parameters that can accurately describe the temporal characteristics of the video were explored and extracted, as well as a homogenization method for fusing heterogeneous feature parameters, and the temporal domain JND model based on this was improved.The feature parameters were investigated including foreground and background motion, temporal duration along the motion trajectory, residual fluctuation intensity along motion trajectory and adjacent inter-frame prediction residual, etc., which were used to characterize the temporal characteristics.Probability density functions for these feature parameters in the perception sense according to the HVS(human visual system) characteristics were proposed, and uniformly mapping the heterogeneous feature parameters to the scales of self-information and information entropy to achieve a homogeneous fusion measurement.The coupling method of visual attention and masking was explored from the perspective of energy distribution, and the temporal-domain JND weight model was constructed accordingly.On the basis of the spatial JND threshold, the temporal domain weights was integrated to develop a more accurate spatial-temporal JND model.In order to evaluate the performance of the spatiotemporal JND model, a subjective quality evaluation experiment was conducted.Experimental results justify the effectiveness of the proposed model.…”
Get full text

A general approach for determining applicability domain of machine learning models by Lane E. Schultz, Yiqi Wang, Ryan Jacobs, Dane Morgan

“…We also show that high measures of dissimilarity are associated with poor model performance (i.e., high residual magnitudes) and poor estimates of model uncertainty (i.e., unreliable uncertainty estimation). …”
Get full text

Impact of transfer learning methods and dataset characteristics on generalization in birdsong classification by Burooj Ghani, Vincent J. Kalkman, Bob Planqué, Willem-Pier Vellinga, Lisa Gill, Dan Stowell

Get full text

Hierarchical Modeling for Medical Visual Question Answering with Cross-Attention Fusion by Junkai Zhang, Bin Li, Shoujun Zhou

“…The hierarchical prompting module pre-aligns hierarchical text prompts with image features to guide the model in focusing on specific image regions according to question types, while the hierarchical decoder performs separate predictions for questions at different levels to improve accuracy across granularities. …”
Get full text