CMStatistics 2021: Start Registration
View Submission - CMStatistics
Title: Combining user-based collaborative filtering and classification for matching footwear size Authors:  Aleix Alcacer - Universitat Jaume I (Spain)
Irene Epifanio - Universitat Jaume I (Spain) [presenting]
Jorge Valero - Instituto de Biomecanica de Valencia (Spain)
Alfredo Ballester - Instituto de Biomecanica de Valencia (Spain)
Abstract: Size mismatch is a serious problem in online footwear purchases because size mismatch implies an almost sure return. Not only foot measurements are important in selecting a size, but also user preference. Therefore, we propose several methodologies that combine the information provided by a classifier with anthropometric measurements and user preference information through user-based collaborative filtering. As novelties: (1) the information sources are 3D foot measurements from a low-cost 3D foot digitizer, past purchases and self-reported size; (2) we propose to use an ordinal classifier after imputing missing data with different options based on the use of collaborative filtering; (3) we also propose an ensemble of ordinal classification and collaborative filtering results; and (4) several methodologies based on clustering and archetype analysis are introduced as user-based collaborative filtering for the first time. The hybrid methodologies were tested in a simulation study, and they were also applied to a dataset of Spanish footwear users. The results show that combining the information from both sources predicts the foot size better, and the new proposals provide better accuracy than the classic alternatives considered.