Dimension Reduction Method by Principal Component Analysis in the Prediction of Final User Satisfaction
Kitti Koonsanit, Daiki Hiruma, Nobuyuki Nishiuchi · 2022 12th International Congress on Advanced Applied Informatics (IIAI-AAI) · 2022
User Experience (UX) issues are widely used for product or service evaluation. Although UX data provide abundant information about behavior log data, feeling and user satisfaction, high dimensionality and long time periods substantially increase the computational burden. In UX evaluation, one challenging task in pre-processing is to reduce the redundancy of the UX data without losing valuable details. In this paper, we present a dimension reduction method by Principal Component Analysis (PCA) in the prediction of final user satisfaction. The dimension reduction method in our research not only serves as the first step in UX data pre-processing that leads to a significant reduction in computational complexity, it is also an invaluable research tool for retaining key information in UX data. In this study, we tested the dimension reduction method with PCA based on satisfaction questionnaires about travel website usage. The proposed method achieves good results in terms of accuracy when compared with original UX data.