Yi-Ting Peng^1,*, Chin-Laung Lei²

CMES-Computer Modeling in Engineering & Sciences, Vol.145, No.3, pp. 3969-3992, 2025, DOI:10.32604/cmes.2025.073258 - 23 December 2025

Abstract The rapid advancement of Large Language Models (LLMs) has enabled their application in diverse professional domains, including law. However, research on automatic judicial document generation remains limited, particularly for Taiwanese courts. This study proposes a keyword-guided training framework that enhances LLMs’ ability to generate structured and semantically coherent judicial decisions in Chinese. The proposed method first employs LLMs to extract representative legal keywords from absolute court judgments. Then it integrates these keywords into Supervised Fine-Tuning (SFT) and Reinforcement Learning with Human Feedback using Proximal Policy Optimization (RLHF-PPO). Experimental evaluations using models such as Chinese Alpaca More >

Muhammad Saad Farooqui¹, Aizaz Ahmad Khattak², Bakri Hossain Awaji³, Nazik Alturki⁴, Noha Alnazzawi⁵, Muhammad Hanif^6,*, Muhammad Shahbaz Khan²

CMES-Computer Modeling in Engineering & Sciences, Vol.145, No.3, pp. 4395-4439, 2025, DOI:10.32604/cmes.2025.072023 - 23 December 2025

Abstract Quality of Service (QoS) assurance in programmable IoT and 5G networks is increasingly threatened by cyberattacks such as Distributed Denial of Service (DDoS), spoofing, and botnet intrusions. This paper presents AutoSHARC, a feedback-driven, explainable intrusion detection framework that integrates Boruta and LightGBM–SHAP feature selection with a lightweight CNN–Attention–GRU classifier. AutoSHARC employs a two-stage feature selection pipeline to identify the most informative features from high-dimensional IoT traffic and reduces 46 features to 30 highly informative ones, followed by post-hoc SHAP-guided retraining to refine feature importance, forming a feedback loop where only the most impactful attributes are More >

Abd Rahman Wahid^*

Journal of Cyber Security, Vol.7, pp. 589-614, 2025, DOI:10.32604/jcs.2025.071622 - 11 December 2025

Abstract The rapid increase in cyber attacks requires accurate, adaptive, and interpretable detection and response mechanisms. Conventional security solutions remain fragmented, leaving gaps that attackers can exploit. This study introduces the HI-XDR (Hybrid Intelligent Extended Detection and Response) framework, which combines network-based Suricata rules and endpoint-based Wazuh rules into a unified dataset containing 45,705 entries encoded into 1058 features. A semantic-aware autoencoder-based anomaly detection module is trained and strengthened through adversarial learning using Projected Gradient Descent, achieving a minimum mean squared error of 0.0015 and detecting 458 anomaly rules at the 99th percentile threshold. A comparative… More >

Ricardo Dominico Da Silva^1,2, Jangkung Raharjo^1,3,*, Sudarmono Sasmono^1,3

Energy Engineering, Vol.122, No.12, pp. 5073-5090, 2025, DOI:10.32604/ee.2025.071545 - 27 November 2025

Abstract The rapid development of technology has led to an ever-increasing demand for electrical energy. In the context of Timor-Leste, which still relies on fossil energy sources with high operational costs and significant environmental impacts, electricity load forecasting is a strategic measure to support the energy transition towards the Net Zero Emission (NZE) target by 2050. This study aims to utilize historical electricity load data for the period 2013–2024, as well as data on external factors affecting electricity consumption, to forecast electricity load in Timor-Leste in the next 10 years (2025–2035). The forecasting results are expected… More >

Zhiwei Ye^1,2,3, Dingfeng Song¹, Haitao Xie^1,2,3,*, Jixin Zhang^1,2, Wen Zhou^1,2, Mengya Lei^1,2, Xiao Zheng^1,2, Jie Sun¹, Jing Zhou¹, Mengxuan Li¹

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5509-5530, 2025, DOI:10.32604/cmc.2025.067342 - 23 October 2025

Abstract The Multilayer Perceptron (MLP) is a fundamental neural network model widely applied in various domains, particularly for lightweight image classification, speech recognition, and natural language processing tasks. Despite its widespread success, training MLPs often encounter significant challenges, including susceptibility to local optima, slow convergence rates, and high sensitivity to initial weight configurations. To address these issues, this paper proposes a Latin Hypercube Opposition-based Elite Variation Artificial Protozoa Optimizer (LOEV-APO), which enhances both global exploration and local exploitation simultaneously. LOEV-APO introduces a hybrid initialization strategy that combines Latin Hypercube Sampling (LHS) with Opposition-Based Learning (OBL), thus… More >

Jiming Lan¹, Hongwei Fu^1,*, Yadong Wu^1,2, Yaxian Liu^1,3, Jianhua Dong^1,2, Wei Liu^1,2, Huaqiang Chen^1,2

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 4645-4665, 2025, DOI:10.32604/cmc.2025.067289 - 23 October 2025

Abstract In the research and production of fluorinated materials, large volumes of unstructured textual data are generated, characterized by high heterogeneity and fragmentation. These issues hinder systematic knowledge integration and efficient utilization. Constructing a knowledge graph for fluorinated materials processing is essential for enabling structured knowledge management and intelligent applications. Among its core components, Named Entity Recognition (NER) plays an essential role, as its accuracy directly impacts relation extraction and semantic modeling, which ultimately affects the knowledge graph construction for fluorinated materials. However, NER in this domain faces challenges such as fuzzy entity boundaries, inconsistent terminology,… More >

Abdulaziz Alshahrani^*

CMC-Computers, Materials & Continua, Vol.85, No.2, pp. 4067-4090, 2025, DOI:10.32604/cmc.2025.065211 - 23 September 2025

Abstract High dropout rates in short-term job skills training programs hinder workforce development. This study applies machine learning to predict program completion while addressing class imbalance challenges. A dataset of 6548 records with 24 demographic, educational, program-specific, and employment-related features was analyzed. Data preprocessing involved cleaning, encoding categorical variables, and balancing the dataset using the Synthetic Minority Oversampling Technique (SMOTE), as only 15.9% of participants were dropouts. six machine learning models—Logistic Regression, Random Forest, Support Vector Machine, K-Nearest Neighbors, Naïve Bayes, and XGBoost—were evaluated on both balanced and unbalanced datasets using an 80-20 train-test split. Performance More >

MELISSA SáNCHEZ-RODRíGUEZ^1,2, ROBERTO LAZZARINI-LECHUGA³, VERóNICA SOUZA-ARROYO^1,4, LETICIA BUCIO-ORTIZ^1,4, ROXANA U. MIRANDA-LABRA^1,4, MONSERRAT GERARDO-RAMíREZ¹, ARACELI PáEZ-ARENAS⁵, MOISES VERGARA-MENDOZA⁶, MARíA CONCEPCIóN GUTIéRREZ-RUIZ^1,4, ALEJANDRO ESCOBEDO-CALVARIO^1,2,*, LUIS E. GOMEZ-QUIROZ^1,4,*

Oncology Research, Vol.33, No.8, pp. 2075-2084, 2025, DOI:10.32604/or.2025.064899 - 18 July 2025

Abstract Background: Growth differentiation factor 11 (GDF11), a transforming growth factor-beta superfamily member, is a crucial protein involved in many differentiation processes in embryogenesis and morphogenesis, and it has been extensively characterized due to its capacity to target poorly differentiated cells, including transformed or cancer cells. Aim: In the present work, we aimed to describe the effects on migration, proliferation, and metabolism in the T-cell acute lymphoblastic leukemia-derived cell line Jurkat. Methods: Based on previous evidence, we analyzed metabolic changes exerted by GDF11 and its relationship with the aggressive phenotype. Results: We found a profound impact on More >

Go-Eun Woo, Sang-Bo Park, Gi-Tae Park, Muhammad Junaid, Hyung-Won Kim^*

CMC-Computers, Materials & Continua, Vol.84, No.2, pp. 3241-3257, 2025, DOI:10.32604/cmc.2025.063723 - 03 July 2025

Abstract On-device Artificial Intelligence (AI) accelerators capable of not only inference but also training neural network models are in increasing demand in the industrial AI field, where frequent retraining is crucial due to frequent production changes. Batch normalization (BN) is fundamental to training convolutional neural networks (CNNs), but its implementation in compact accelerator chips remains challenging due to computational complexity, particularly in calculating statistical parameters and gradients across mini-batches. Existing accelerator architectures either compromise the training accuracy of CNNs through approximations or require substantial computational resources, limiting their practical deployment. We present a hardware-optimized BN accelerator… More >

Adéla Hamplová^*, Tomáš Novák, Miroslav Žáček, Jiří Brožek

CMES-Computer Modeling in Engineering & Sciences, Vol.143, No.3, pp. 3329-3349, 2025, DOI:10.32604/cmes.2025.066025 - 30 June 2025

Abstract This study proposes a new component of the composite loss function minimised during training of the Super-Resolution (SR) algorithms—the normalised structural similarity index loss , which has the potential to improve the natural appearance of reconstructed images. Deep learning-based super-resolution (SR) algorithms reconstruct high-resolution images from low-resolution inputs, offering a practical means to enhance image quality without requiring superior imaging hardware, which is particularly important in medical applications where diagnostic accuracy is critical. Although recent SR methods employing convolutional and generative adversarial networks achieve high pixel fidelity, visual artefacts may persist, making the design of… More >