Abstract High-dimensional sensor data present substantial analytical challenges due to feature sparsity, nonlinear dependencies, noisy observations, and unstable clustering behavior in conventional machine learning frameworks. Although deep representation learning and subspace clustering methods have significantly advanced high-dimensional analytics, many existing approaches remain fragmented, operating independently across latent representation learning, semantic embedding optimization, and clustering stages. This study proposes a unified multi-stage unsupervised sensor intelligence framework integrating... High-dimensional sensor data present substantial analytical challenges due to feature sparsity, nonlinear dependencies, noisy observations, and unstable clustering behavior in conventional machine learning frameworks. Although deep representation learning and subspace clustering methods have significantly advanced high-dimensional analytics, many existing approaches remain fragmented, operating independently across latent representation learning, semantic embedding optimization, and clustering stages. This study proposes a unified multi-stage unsupervised sensor intelligence framework integrating Variational Autoencoder (VAE)-based probabilistic latent representation learning, contrastive representation optimization using Triplet Networks, SimCLR, and BYOL, and consensus hybrid clustering for robust subgroup discovery in high-dimensional digital sensor signals. The proposed framework is evaluated using the IEEE DataPort binary-classification digital sensor signal dataset comprising 11 numerical sensor features characterized by temporal dependencies and event-driven activation patterns. Experimental evaluation incorporates quantitative clustering metrics including Silhouette Score, Davies–Bouldin Index, Calinski– Harabasz Index, reconstruction analysis, cluster stability evaluation, comparative baselines, and consensus clustering validation. Results demonstrate that probabilistic latent manifold learning substantially improves structural organization, semantic continuity, and anomaly-sensitive clustering compared with conventional dimensionality reduction and raw-feature clustering approaches. Hierarchical clustering provides balanced subgroup segmentation, whereas spectral clustering exhibits strong sensitivity toward sparse anomalous structures. The proposed consensus hybrid clustering mechanism combines these complementary strengths to produce stable and semantically coherent latent partitions. The study contributes a robust and scalableframework for intelligent sensing, anomaly detection, cyber-physical monitoring, and unsupervised IoT analytics. Read More Read Less

Abstract The rapid integration of artificial intelligence (AI) into global labor markets is fundamentally reshaping occupational structures and workforce dynamics. This study proposes an Explainable AI-Driven Workforce Intelligence Framework for automation risk analysis and occupational transformation toward 2030. Addressing critical gaps in interpretability within existing workforce-automation research, the framework integrates predictive analytics, global and local explainability mechanisms, and attention-inspired feature interdependency analysis to investigate how socioeconomic... The rapid integration of artificial intelligence (AI) into global labor markets is fundamentally reshaping occupational structures and workforce dynamics. This study proposes an Explainable AI-Driven Workforce Intelligence Framework for automation risk analysis and occupational transformation toward 2030. Addressing critical gaps in interpretability within existing workforce-automation research, the framework integrates predictive analytics, global and local explainability mechanisms, and attention-inspired feature interdependency analysis to investigate how socioeconomic and technological variables collectively influence occupational vulnerability. Using the AI Impact on Jobs 2030 dataset, a Gradient Boosting-based predictive model classifies occupations into risk categories, while permutation-based feature attribution and Local Interpretable Model-Agnostic Explanations (LIME) provide transparent interpretation of workforce-risk determinants. Results reveal that automation feasibility indicators exert substantially stronger influence on risk prediction than traditional socioeconomic variables, with Automation Probability 2030 emerging as the dominant predictive anchor. Attention-inspired dependency analysis further demonstrates that high-risk occupations exhibit intensified feature interdependencies, particularly between technological growth and workforce experience, suggesting that rapid technological advancement may diminish the stabilizing influence of professional tenure. The study contributes an integrated, interpretable analytical architecture that extends explainability concepts from natural language processing to workforce intelligence, enabling transparent occupational-risk assessment and evidence-driven labor-market forecasting. By illuminating the structural mechanisms underlying workforce vulnerability, the framework supports policymakers, educational institutions, and organizations in designing targeted reskilling initiatives and adaptive workforce strategies for sustainable labor-market transformation in the era of AI-driven digital change. Read More Read Less

Abstract Radio frequency (RF) signal datasets exhibit complex nonlinear distributions, environmental noise, and heterogeneous transmission behaviors that challenge traditional clustering algorithms. Conventional ensemble approaches often rely on static weighting strategies and overlook local structural uncertainty, limiting their robustness in dynamic wireless environments. To address these limitations, this study introduces an Adaptive Weighted Ensemble Clustering framework designed for robust latent structure discovery in RF signal datasets. The... Radio frequency (RF) signal datasets exhibit complex nonlinear distributions, environmental noise, and heterogeneous transmission behaviors that challenge traditional clustering algorithms. Conventional ensemble approaches often rely on static weighting strategies and overlook local structural uncertainty, limiting their robustness in dynamic wireless environments. To address these limitations, this study introduces an Adaptive Weighted Ensemble Clustering framework designed for robust latent structure discovery in RF signal datasets. The proposed methodology integrates multiple clustering algorithms, including centroidbased, density aware, hierarchical, and graph based methods, into a unified consensus architecture. By constructing a performance-driven weighted co-association matrix and incorporating local uncertainty estimation, the framework dynamically optimizes algorithmic contributions while suppressing unstable partitions. High order topological enhancement further refines neighborhood relationships, ensuring coherent subgroup formation in sparse spectral regions. Experimental evaluation demonstrates that the adaptive consensus model significantly outperforms standalone clustering baselines, achieving a consensus silhouette score of 0.722 and exhibiting exceptional reproducibility across subsampling trials. Co-association heatmap analysis reveals distinct block-diagonal structures, confirming the reliable identification of latent transmission subgroups. The framework’s dynamic adaptation capability ensures sustained performance under varying signal to noise ratios and heterogeneous dataset distributions. By bridging the gap between algorithmic instability and the stringent reliability requirements of modern RF analytics, this work establishes a scalable foundation for unsupervised spectrum intelligence, anomaly detection, and AI-driven wireless security. Future research will integrate self supervised representation learning and optimize computational efficiency for real-time edge deployment. Read More Read Less