Window-Based Feature Engineering for Cognitive Workload Detection

2025-11-02

AI Summary

This study addresses the challenge of real-time, multi-level cognitive workload (CW) assessment in complex dynamic tasks. We propose a sliding-window-based temporal feature enhancement method and conduct physiological-signal-driven CW classification using the COLET dataset. Methodologically, we systematically compare handcrafted feature extraction combined with conventional machine learning models (e.g., SVM, Random Forest) against deep learning architectures tailored for tabular time-series data—specifically TabNet—thereby providing the first empirical validation of TabNet’s efficacy on physiological tabular time-series data. Results demonstrate that our windowed feature engineering significantly improves model robustness; TabNet consistently outperforms traditional approaches across accuracy, precision, and F1-score (average gains of 3.2–5.8 percentage points), enabling more stable and accurate real-time multi-level CW recognition. This work establishes a deployable modeling paradigm for wearable neuroergonomics applications.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Hardware-aware ML
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Cognitive workload is a topic of increasing interest across various fields such as health, psychology, and defense applications. In this research, we focus on classifying cognitive workload using the COLET dataset, employing a window-based approach for feature generation and machine/deep learning techniques for classification. We apply window-based temporal partitioning to enhance features used in existing research, followed by machine learning and deep learning models to classify different levels of cognitive workload. The results demonstrate that deep learning models, particularly tabular architectures, outperformed traditional machine learning methods in precision, F1-score, accuracy, and classification precision. This study highlights the effectiveness of window-based temporal feature extraction and the potential of deep learning techniques for real-time cognitive workload assessment in complex and dynamic tasks.

Problem

• Classifying cognitive workload levels using window-based feature engineering
• Enhancing temporal feature extraction through window-based partitioning methods
• Comparing deep learning and traditional machine learning classification performance

Innovation

• Window-based temporal partitioning for feature enhancement
• Deep learning models outperforming traditional machine learning
• Tabular architectures for real-time cognitive workload assessment