Machine Learning Technique for Data Fusion and Cognitive Load Classification Using an Eye Tracker

Document Type

Conference Proceeding

Publication Date

1-1-2023

Publication Title

Lecture Notes in Networks and Systems

Volume

700 LNNS

First Page

86

Keywords

cognitive load, eye-tracker, human-machine automation, Machine Learning, n-back tasks

Last Page

95

Abstract

Advanced Driver Assistance Systems (ADAS) and other human-machine automation systems should be able to accurately recognize and adapt to the cognitive load of the user. For effective human-machine automation, it is important to develop techniques to automatically predict the cognitive load, based on data from non-invasive and low-cost sensors, such as eye-trackers. In this paper, we investigate the use of machine learning (ML) to classify the cognitive load of a participant performing n-back tasks. The ML models are trained using a large number of raw eye-tracking metrics. Our results demonstrate that tree-based algorithms are able to quickly predict cognitive load with a high degree of accuracy compared to other methods, indicating their potential usefulness for real-time applications.

DOI

10.1007/978-3-031-33743-7_7

ISSN

23673370

E-ISSN

23673389

ISBN

9783031337420

Share

COinS