Research on Time Pressure Recognition of Online Shopping Users Based on Eye Movement Data
[Purpose/significance]This paper aims to establish a machine learning model based on eye movement data to identify the time pressure level of users when shopping.[Method/process]The paper recruits a total of 32 subjects to conduct an eye tracking experiment on shopping.The subjects perform four tasks under different levels of time pressure,select machine learning algorithms such as random forest,support vector machine,gradient boosting tree and k-nearest neighbor to construct the recognition model,use the accuracy rate,recall rate,precision rate,the values of F1 and ROC index assessment model.[Result/conclusion]Random forest has the best recognition accuracy,and the prediction accuracy on the test set reaches 87.5％,in which the fixation eye movement indicators such as fixation duration and fixation number contribute the most to the recognition model.
shopping behaviortime pressureeye movement experimentmachine learning
万方数据
维普
