Jihoon Moon^*

CMC-Computers, Materials & Continua, Vol.89, No.2, 2026, DOI:10.32604/cmc.2026.089115 - 15 September 2026

Abstract Large language models and retrieval-augmented generation (RAG) systems are increasingly employed to transform evidence into decision-facing briefings, alerts, and recommendations. In these settings, explainability cannot be evaluated merely by fluency, readability, or factual correctness. A briefing may be factually correct while still being unsafe if it cites sources that do not substantiate the claim, suppresses uncertainty, converts correlational evidence into causal language, recommends an unauthorized action, or leaves no auditable path for human review. This review synthesizes 104 sources spanning explainable natural language processing (NLP), faithful explanation, hallucination and factuality evaluation, RAG, citation faithfulness, uncertainty… More >

Engin Seven^1,*, Eylem Yucel¹, Munevver Yildirim²

CMC-Computers, Materials & Continua, Vol.89, No.2, 2026, DOI:10.32604/cmc.2026.083509 - 15 September 2026

Abstract Objective, interpretable decision support for Post-Traumatic Stress Disorder (PTSD) screening remains a challenge in computational psychiatry, where existing methods either rely on costly neuroimaging or lack the diagnostic transparency required for clinical accountability. This study presents Multimodal HetGAT-PTSD, a heterogeneous graph attention network (HetGAT) that integrates unstructured clinical narratives with structured item-level responses from the PTSD Checklist for DSM-5 (PCL-5). The model operates under a graph topology constrained by the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria to ensure structural alignment between clinical theory and graph-based learning. For each patient, a 25-node directed… More >

Wen-Chin Hsu, Yi-Cheng Chen^*, Yi-Hsuan Kuo

CMES-Computer Modeling in Engineering & Sciences, Vol.148, No.2, 2026, DOI:10.32604/cmes.2026.085408 - 28 August 2026

Abstract Due to the structural complexity of Chinese characters, the occurrence of homophones and visual similarity among glyphs directly increases the difficulties presented in Chinese spell checking (CSC). These factors also indicate the importance of the connection between CSC and context-dependency. In this study, a novel framework, the Summarization-Enhanced Chinese Spell Checking (abbreviated as SE-CSC) model, is proposed, which integrates phonetic and glyph embeddings to further enhance context awareness in error detection and correction. We utilize sentence-level summarization features to augment and generate an error-guided mask that can effectively detect errors and derive more precise corrections. More >

Mrugendrasinh Rahevar¹, Martin Parmar¹, Hemant Yadav¹, Chun-Ta Li^2,*, Agbotiname Lucky Imoize³, Hiren Mewada⁴

CMES-Computer Modeling in Engineering & Sciences, Vol.148, No.2, 2026, DOI:10.32604/cmes.2026.085393 - 28 August 2026

Abstract Quantum kernel techniques aim to leverage quantum computational capabilities on social data. However, their application to natural language processing tasks faces formidable obstacles, such as extreme dimensionality reduction ($D=384\to k=8$), concentration of measure in quantum feature spaces, and the lack of theoretical understanding of when quantum advantages occur in kernel-based text classification. Filling this gap, we provide a comprehensive study of quantum kernels for text classification that addresses three major challenges in existing studies: general data compression approaches that ignore class structure, the lack of a predictive diagnostic toolkit, and overlooked approaches for handling concentration… More >

Amine Gonca Toprak^1,*, Aytuğ Onan²

CMC-Computers, Materials & Continua, Vol.89, No.1, 2026, DOI:10.32604/cmc.2026.085268 - 13 August 2026

Abstract Biomedical texts present significant challenges for natural language processing (NLP) due to their complex terminology, intricate contextual dependencies, and highly domain-specific semantics. This study investigates the effectiveness of knowledge distillation (KD) for biomedical text classification, aiming to develop lightweight, resource-efficient models that remain competitive with larger architectures. A balanced dataset of 25,000 PubMed records was constructed, equally distributed across five biomedical domains. Two teacher models (BERT and PubMedBERT) and five student models (DistilBERT, BioClinicalBERT, BioBERT, DistilBioBERT, and DistilRoBERTa) were evaluated across ten distinct KD configurations. Each student model was also directly fine-tuned to serve as… More >

Shunran Duan, Meijuan Yin^*, Xiangyang Luo, Lunchong Cui, Chenyu Wang

CMC-Computers, Materials & Continua, Vol.89, No.1, 2026, DOI:10.32604/cmc.2026.084290 - 13 August 2026

Abstract Few-shot relation classification aims to identify semantic relations between entity pairs under limited annotated data. Although recent prompt learning-based methods have achieved promising performance, they often rely on manually crafted, domain-specific prompt templates, which restrict their transferability across domains. In this paper, building upon the multi-task prompt transfer paradigm of MPT, we propose a Confidence-guided Orthogonality-constraint Prompt Adaptation framework for few-shot relation classification, named COPA. The proposed framework learns a shared domain-invariant prompt matrix together with domain-specific low-rank prompt matrices via multi-domain soft prompt tuning, enabling the transfer of domain-invariant relational knowledge across domains. Unlike… More >

Laura Baitenova¹, Gulnar Mukhamejanova², Gauhar Munaitbas^3,*, Saken Mambetov¹, Zhanna Mukanova¹

CMC-Computers, Materials & Continua, Vol.88, No.3, 2026, DOI:10.32604/cmc.2026.081004 - 23 July 2026

Abstract Morphological parsing is a fundamental task in natural language processing, particularly for morphologically rich languages where words encode complex grammatical and semantic information. This paper proposes a multi-branch Transformer-enhanced neural framework for joint morphological representation learning, designed to improve segmentation and classification accuracy by integrating complementary feature extraction mechanisms. The proposed architecture combines convolutional layers for capturing local morphological patterns, recurrent layers for modeling sequential dependencies, and Transformer-based self-attention for learning global contextual relationships. This hybrid design enables the model to generate robust and context-aware representations that enhance morphological understanding. The framework is trained using… More >

Naimul Hasan Shadesh^*, Zannatul Ferdous, Bipasha Iasmin

Journal on Artificial Intelligence, Vol.8, pp. 335-357, 2026, DOI:10.32604/jai.2026.078589 - 22 July 2026

Abstract Sentiment analysis is a core task in Natural Language Processing (NLP) that aims to identify opinions and sentiment polarity expressed in textual data. This study presents a systematic empirical evaluation of classical machine learning–based sentiment analysis methods using a unified experimental framework. Several supervised classifiers, including Decision Trees, Logistic Regression, Support Vector Machines (SVM), Random Forests, Naïve Bayes, and K-Nearest Neighbors (KNN), are evaluated on labeled text datasets collected from multiple domains such as product reviews, customer feedback, hotel reviews, and social media content. The experimental pipeline includes standard NLP preprocessing steps—text normalization, tokenization, stopword More >

Ghadi Almaktoom, Suliman Aladhadh, Salim El Khediri^*

CMES-Computer Modeling in Engineering & Sciences, Vol.147, No.2, 2026, DOI:10.32604/cmes.2026.078774 - 27 May 2026

Abstract The rapid growth of phishing attempts in the enterprise could potentially lead to bankruptcy. The primary focus of the research is on detecting phishing attacks, with no interest in how the data is processed. Attackers use fraudulent methods to obtain valuable, confidential information, resulting in billions of dollars in financial losses for enterprises. In our review, we examined the methods used in phishing-detection studies. We concluded that the two main sections, centralized and decentralized methods, were the centralized ones, which aggregate data in a central server and thus violate data protection regulations, such as GDPR.… More >

Choongwon Kang^1,2, Haein Lee^3,4, Jang Hyun Kim^1,2,*

CMC-Computers, Materials & Continua, Vol.88, No.1, 2026, DOI:10.32604/cmc.2026.078997 - 08 May 2026

Abstract With the rapid growth of social media and frequent anonymous interactions, hate speech has become widespread. As users express diverse opinions in digital spaces, the need for effective detection remains crucial. To address this, we propose a framework applicable to diverse hate speech types, combining sentence-level semantic representation vectors from the pre-trained Bidirectional Encoder Representations from Transformers (BERT) with sentiment score vectors from the Linguistic Inquiry and Word Count (LIWC) dictionary and the Valence Aware Dictionary for sEntiment Reasoning (VADER). This semantic-sentiment fusion integrates three deep learning models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), More >