A Comparison of AutoML Hyperparameter Optimization Tools For Tabular Data by Prativa Pokhrel, Alina Lazar

“…Therefore, finding the optimal values of these hyperparameters is integral in improving the prediction accuracy of an ML algorithm and the model selection. …”
Get full text

Exploring Gender Differences in Fatwa through Machine Learning by Emad Mohamed, Raheem Sarwar

“…Our methodology involves comprehensive experimental analysis, utilizing advanced Arabic text preprocessing alongside machine learning algorithms. We concentrate on two primary objectives: predicting the gender of the questioner and forecasting the popularity of answers. …”
Get full text

Energy efficient task scheduling for heterogeneous multicore processors in edge computing by Yanchun Liu, Hongxue Qu, Shuang Chen, Xuejun Feng

“…This paper proposes a novel algorithm integrating task prioritization, core-aware mapping, and predictive DVFS. …”
Get full text

Deep neural networks have an inbuilt Occam’s razor by Chris Mingard, Henry Rees, Guillermo Valle-Pérez, Ard A. Louis

“…Combining this with the prior yields an accurate prediction for the posterior, measured for DNNs trained with stochastic gradient descent. …”
Get full text

Soft Sensor Modeling of Key Effluent Parameters in Wastewater Treatment Process Based on SAE-NN by Yousuf Babiker M. Osman, Wei Li

“…A soft sensor model is studied to predict the BOD5 in a wastewater treatment plant to evaluate the proposed approach. …”
Get full text

A multi-chroma format cascaded coding method for full-chroma image in AVS2 by Liping ZHAO, Tao LIN, Kailun ZHOU, Keli HU, Chunmei LIN

“…In the second generation of audio video coding standard (AVS2),a multi-chroma format cascaded coding method (MCFCC) for full-chroma (4:4:4 sampling format) images coding was proposed.The MCFCC algorithm firstly converts the 4:4:4 sampling format image into 4:2:0 sampling format image,then the 4:2:0 sampling format image was processed by 4:2:0 sampling format intra prediction,transform,quantization,inverse quantization,inverse transform,entropy coding and a weighted 4:4:4 sampling format distortion calculation method in the rate-distortion optimization process,finally 4:4:4 sampling format in-loop filtering and offset algorithms were applied to the 4:4:4 sampling format image after up sampling.The experimental results show that,for full-chroma natural images,the MCFCC algorithm achieves higher coding efficiency at very low additional encoding complexity and very low additional design and implementation cost.…”
Get full text

Modularized Learning for Hate Speech Detection in Korean: Integrating Emotions and Multi-Faceted Attributes by Hyeun Jeong Min

“…Built on pre-trained Korean language models, our framework extracts features related to emotion, multiple attributes, and similarity, and then employs a classifier for final prediction. Experimental results demonstrate that our proposed method outperforms existing algorithms by achieving a higher f1-score in hate speech recognition.…”
Get full text

Edge of excellence: JEC's Scopus achievement, doors 2025 breakthroughs, and Volume 4 Issue 1 (2025) highlights by Serhiy O. Semerikov, Tetiana A. Vakaliuk

“…The paper examines innovative solutions addressing distributed systems challenges, including IoT-based monitoring systems, intelligent traffic prediction algorithms, and security implementations for resource-constrained environments. …”
Get full text

Dual branch guided contrastive learning for unsupervised pedestrian re-identification by REN Hangjia, LIANG Fengmei

“…The current unsupervised pedestrian re-identification algorithms using residual networks can only extract rough global features, but it can’t adequately reflect subtle local features. …”
Get full text

Exposing Face Manipulation Based on Generative Adversarial Network–Transformer and Fake Frequency Noise Traces by Qiaoyue Man, Young-Im Cho

“…In recent years, with the application of GANs and diffusion generative network algorithms, many highly realistic synthetic images are emerging, greatly increasing the potential for misuse, and deepfakes have become a serious social concern. …”
Get full text

The use of web resources for metabolomics in horticultural crops by Esra Karakas, Mustafa Bulut, Alisdair R. Fernie

“…Moreover, the application of machine learning algorithms to these web resources has further optimized data interpretation, enabling more accurate prediction of metabolic profiles. …”
Get full text

A Method for Aileron Actuator Fault Diagnosis Based on PCA and PGC-SVM by Wei-Li Qin, Wen-Jin Zhang, Chen Lu

“…Second, the data feature extraction is performed with dimension reduction using PCA. Finally, the GS and CV algorithms are employed to train a one-versus-one SVM for fault classification, thus obtaining the optimal model parameters and assuring the generalization of the trained SVM, respectively. …”
Get full text

Signal-piloted processing and machine learning based efficient power quality disturbances recognition. by Saeed Mian Qaisar

“…Therefore, a remarkable reduction can be secured in the data storage, processing and transmission requirement towards the post classifier. …”
Get full text

Research and application of natural air-cooled traction transformer for EMU by JIN Jun

“…Verification results show that, compared with traditional forced air-cooled traction transformers for EMUs, this new natural air-cooled traction transformer achieves a 2% improvement in efficiency, a 50% reduction in losses, a decrease in noise levels by 15 dB(A), a reduction of approximately 90% in the maintenance workload for the cooling system, and a decrease in total lifecycle costs of about 40%. …”
Get full text

Analog Circuits Fault Diagnosis Using ISM Technique and a GA-SVM Classifier Approach by Sabah Kouachi, Nacerdine Bourouba, Kamel Mebarkia, Imad Laidani

“…This is built with GA and SVM algorithms combination targeting both data reduction and fault classification accuracy respectively. …”
Get full text

Dynamic strategy for adaptive block size optimization in blockchain technology by Shafique Ahmed Awan, Muazzam Ali Khan Khattak, Anwar Ali Sathio, Haleem Farman, Sumaira Memon, Moustafa M. Nasralla, Ahmed Sedik

“…This study proposes a dynamic block size optimization framework for private blockchain networks using hybrid heuristic algorithms—Whale Optimization Algorithm (WOA) and Particle Swarm Optimization (PSO)—integrated with Merkle trees and Directed Acyclic Graphs (DAGs). …”
Get full text

Assessing the effectiveness of transfer learning strategies in BLSTM networks for speech denoising by Marvin Coto-Jiménez, Astryd González-Salazar, Michelle Gutiérrez-Muñoz

“…Results show the convenience of transferring information from a single case of denoising network to the rest, with a significant reduction in training time and denoising capabilities of the BLSTM networks. …”
Get full text

Facial emotion based smartphone addiction detection and prevention using deep learning and video based learning by C. Joseph, P. Uma Maheswari

“…Experimental results demonstrate significant improvements in students’ behavior and reductions in smartphone usage post-intervention. The TMVM system achieves high accuracy in emotion detection and behavioral outcome prediction while fostering engagement in school and social activities. . …”
Get full text

Stratified allocation method for water injection based on machine learning: A case study of the Bohai A oil and gas field by Changlong Liu, Pingli Liu, Qiang Wang, Lu Zhang, Zechao Huang, Yuande Xu, Shaojiu Jiang, Le Zhang, Changxiao Cao

“…Second, the training and prediction effects of three machine learning prediction models—support vector machine, BP neural network, and random forest—were compared, and the BP neural network was selected as the machine learning mathematical model for injection allocation optimization. …”
Get full text

Rapid diagnosis of bacterial vaginosis using machine-learning-assisted surface-enhanced Raman spectroscopy of human vaginal fluids by Xin-Ru Wen, Jia-Wei Tang, Jie Chen, Hui-Min Chen, Muhammad Usman, Quan Yuan, Yu-Rong Tang, Yu-Dong Zhang, Hui-Jin Chen, Liang Wang

“…Multiple ML models were constructed and optimized, with the convolutional neural network (CNN) model achieving the highest prediction accuracy at 99%. Gradient-weighted class activation mapping (Grad-CAM) was used to highlight important regions in the images for prediction. …”
Get full text