Classifying Mental Workload Levels Using Semi-Supervised Learning Technique

Abstract

Real-time monitoring and analysis of human operator’s mental workload (MWL) is crucial for development of adaptive/intelligent human-machine cooperative systems in various safety/mission-critical application fields. Although data-driven machine learning (ML) approach has shown promise in MWL recognition, it is usually difficult to acquire sufficient labeled data to train the ML model. This paper proposes semi-supervised extreme learning machines (SS-ELM) for MWL pattern classification using solely a small number of labeled data. The experimental data analysis results are presented to show the effectiveness of the proposed SS-ELM paradigm to effectively exploit a large number of unlabeled data for the real-world 3- or 4-class MWL classification problem.