Author(s): |
Tong Shang, The International WIC Institute ,Beijing University Of Technology, beijing, China;School of Electronic Information and Control Engineering, Beijing University Of Technology, beijing, China Yu Zhou, The International WIC Institute ,Beijing University Of Technology, beijing, China; School of Electronic Information and Control Engineering, Beijing University Of Technology, beijing, China Zhao Liu, The International WIC Institute ,Beijing University Of Technology, beijing, China Xue Tan, The International WIC Institute ,Beijing University Of Technology, beijing, China |
Abstract: |
In this paper, we use
eye tracking to study cognitive mechanisms differences between value extraction
and relation comparison during people’s reading of statistical charts, the main
contents include the following two aspects: First, we explore the behavior
differences between value extraction and relation comparison during people’s
comprehending of statistical charts from the human cognitive behavior; Second,
we conduct pattern classification to eye movement data using support vector
machine (SVM) technology, and further analyze the differences of human thinking
mode during the comprehending of statistical charts. The results show that: (1)
The eye movement indexes such as pupil diameter, mean fixation duration,
fixation duration had no significant difference between value extraction and
relation comparison , while saccade distance had significant differences; (2)
Normalization of data before classification can improve the classification
accuracy; (3) We can get a stable classification accuracy using 5-fold
cross-validation method; (4) The eye movement modes of value extraction and
relation comparison have significant differences, namely, there are significant
cognitive differences between value extraction and relation comparison in graph
comprehension.
|