[3Xin2] Poster session 1

[3Xin2-43] A Note on Classification of Expert-novice Level using motion and gaze data via Spatial Temporal Attention GCN

〇Tatsuki Seino¹, Naoki Saito², Takahiro Ogawa³, Satoshi Asamizu⁴, Miki Haseyama³ (1.Graduate School of Information Science and Technology, Hokkaido University, 2.Office of Institutional Research, Hokkaido University, 3.Faculty of Information Science and Technology , Hokkaido University, 4.National Institute of Technology, Kushiro College)

Keywords:Expert-novice classification, Biometric data, Graph Convolutional Network

This paper presents a method for classifying expert-novice levels using motion and gaze data. The classification of expert-novice levels is crucial for transferring skills from experts to novices, and an improvement in the classification performance is anticipated. The proposed method employs a Spatial-Temporal Attention Graph Convolutional Network that utilizes motion and gaze data for classifying expert-novice levels considering the temporal information of each data type. By utilizing various types of data, the proposed method enhances the performance of the expert-novice classification compared to methods that use a single type of data. In this paper, we conducted expert-novice classification experiments using motion and gaze data collected during a soccer play to assess the effectiveness of simultaneously using different types of data for expert-novice classification.