计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (11): 39-42.

• 研究、探讨 • 上一篇    下一篇

基于SVM的眼动轨迹解读思维状态的研究

陈俊杰,严会霞,相 洁   

  1. 太原理工大学 计算机与软件学院,太原 030024
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-04-11 发布日期:2011-04-11

Study of decoding mental state based on eye tracks using SVM

CHEN JunJie,YAN Huixia,XIANG Jie   

  1. College of Computer and Software,Taiyuan University of Technology,Taiyuan 030024,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-04-11 Published:2011-04-11

摘要: 从人们的眼动轨迹来解读人的思维状态已成为应用心理学的研究热点。以四方趣题为研究材料,通过Tobii眼动仪记录被试解题时的眼动轨迹,以眼动轨迹数据中各个AOI的注视持续时间和回视等综合指标的加权值作为特征值训练SVM分类器。经过三种不同的分类任务的实验验证,SVM分类的准确率很高,泛化能力很强,可以作为眼动轨迹分析的分类方法,从而可根据眼动轨迹解读被试解题时的不同策略。

关键词: SVM分类, 眼动轨迹, 四方趣题, 眼动指标

Abstract: It is a hot topic in applied psychology that interpreting people’s thinking from eye tracks.In this paper,research material is 4*4 sudoku.The Tobii eye tracker records subjects’ eye tracks while solving 4*4 sudouk.And then this paper uses feature values,the weighted value of the composite indicators,such as the duration,regression time of each AOI,to train SVM classifier.The experiments of three different classification tasks show that the classification accuracy is very high,and the ability of generalization is strong.SVM can be used for the classification of eye tracks.So this paper can interpret different problem-solving strategies according to the results of classification of eye tracks.

Key words: Support Vector Machine(SVM) classification, eye track, 4*4 sudoku, eye parameter