Information

30 pages, 5794 KB

Open AccessArticle

NS-Dep-KAN: An Explainable Neuro-Symbolic Framework with Kolmogorov–Arnold Networks for DSM-Guided Depression Assessment

by Qiong Hong, Lailatul Qadri Zakaria and Sabrina Tiun

Information 2026, 17(6), 516; https://doi.org/10.3390/info17060516 - 22 May 2026

Abstract

Automated depression assessment is critical for scalable mental healthcare but faces dual challenges: the lack of clinical interpretability in “black-box” deep learning models and the excessive computational cost of large-scale fusion architectures. To bridge this gap, we propose NS-Dep-KAN, a novel neuro-symbolic framework that harmonizes DSM-5-guided reasoning with Kolmogorov–Arnold Networks (KANs). Our approach leverages a Large Language Model (LLM) to extract symbolic symptom evidence aligned with diagnostic criteria, which then guides the aggregation of multimodal features from frozen pretrained encoders (WavLM and Qwen). Unlike traditional Multi-Layer Perceptrons, the proposed KAN prediction head employs learnable B-spline activation functions to capture complex nonlinear symptom–severity mappings with extreme parameter efficiency. Evaluations on the DAIC-WOZ benchmark demonstrate that NS-Dep-KAN achieves state-of-the-art performance among audio-text models (MAE 2.69, 13.5% improvement over the three-modality baseline MSGAF at MAE 3.11), with only ∼4.9 K trainable parameters. Moreover, the framework offers inherent interpretability, revealing granular symptom contribution profiles that align with clinical intuition. This work establishes a path toward explainable trustworthy AI for mental health screening. Full article

(This article belongs to the Special Issue Advanced AI and Data-Driven Learning Methods for Healthcare Applications)

► Show Figures

Graphical abstract

30 pages, 1675 KB

Open AccessArticle

Predicting Academic Award Recognition Across Disciplines Using Publication-Based Bibliometric Indices and SHAP-Driven Explainability

by Muhammad Shaban Qabil, Hafiza Zarafshan Mukhtiar, Ghulam Mustafa, Muhammad Tanvir Afzal, Isabel De la Torre Díez, Elizabeth Caro Montero and Mirtha Silvana Garat de Marin

Information 2026, 17(6), 515; https://doi.org/10.3390/info17060515 - 22 May 2026

Abstract

Researcher evaluation underpins critical academic decisions, yet traditional bibliometric indicators lack predictive capability and cross-domain generalizability, while most predictive approaches offer limited interpretability and narrow domain validation. This study proposes a SHAP interpretable, multi-domain supervised learning framework for predicting academic award recognition using thirty two publication count-based bibliometric indices. A balanced dataset was constructed across four disciplines, namely Computer Science, Neuroscience, Mathematics, and Civil Engineering, comprising verified awardees from recognized professional societies and matched non-awardee researchers. Eight classifiers were evaluated under stratified five fold cross validation, assessed via accuracy, precision, recall, F1-score, and ROC AUC. The framework achieved domain-specific F1-scores of 0.70 in Computer Science, 0.73 in Neuroscience, 0.72 in Civil Engineering, and 0.78 in Mathematics, with SVM and XGBoost demonstrating the strongest cross-domain robustness across disciplines. SHAP analysis consistently identified normalized h index, h2 family, q2 index, and g index as dominant cross-domain predictors, while domain-specific indicators, including Rm and w indices in Neuroscience and P index in Civil Engineering, reflected disciplinary recognition patterns. By unifying publication-based feature engineering, multi-domain classification, and SHAP explainability within a single reproducible pipeline, this framework offers a scalable, transparent, and evidence-based tool for institutional researcher evaluation. Full article

(This article belongs to the Special Issue Machine Learning for Predictive Analytics: Models, Applications, and Challenges)

► Show Figures

Figure 1

18 pages, 8322 KB

Open AccessArticle

V2W-LLM: Automated Vulnerability to Weakness Mapping Based on Large Language Model

by Ziguo Wang, Mei Nian, Yaling Jing and Jun Zhang

Information 2026, 17(6), 513; https://doi.org/10.3390/info17060513 - 22 May 2026

Abstract

To address the rapid growth of software vulnerabilities, the latency of manual expert classification, and the limitations of existing methods restricted to fixed categories, this paper proposes V2W-LLM, an automated vulnerability-to-weakness mapping model based on Large Language Models (LLMs). First, a dataset of CVE-CWE description pairs is constructed based on established expert correlations from MITRE. Subsequently, the LLM is instruction-tuned on this dataset to leverage its reasoning capabilities in generating CWE-style descriptive text for newly disclosed, unmapped vulnerabilities. Finally, using a BAAI-based embedding model, the semantic representations of the generated text and official CWE descriptions are computed to identify the optimal mapping via cosine similarity (Top-1). Experimental results indicate that V2W-LLM achieves an accuracy of 90.18% and a Macro-F1 of 87.64% in common categories. Furthermore, on the public ChatGPT-VDMEval and the latest 2024 NVD datasets, the model attains F1 scores of 86.02% and 94.02% respectively, validating its effectiveness in automating the vulnerability-to-weakness mapping process. Full article

(This article belongs to the Topic New Trends in Cybersecurity and Data Privacy)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Information, Volume 17, Issue 6 (June 2026) – 3 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI