Basic Consideration of Rough Set C-Means Clustering with Missing Value Processing and Its Application to Collaborative Filtering

Abstract

In clustering-based collaborative filtering (CF), clusters of users with similar preference patterns are extracted, and highly preferred items withinthese clusters are recommended. Since the data used in CF tasks contain uncertainties due to human sensitivities, rough clustering based onrough set theory, which handles these uncertainties, is considered effective. Thus, RSCM-CF, a CF method based on rough set C-means (RSCM)clustering, a type of rough clustering, has been proposed. In this study, we propose RSCM-PDS, an RSCM method incorporating missing valueprocessing using the partial distance strategy (PDS) and examine its application to collaborative filtering as RSCM-PDS-CF. In RSCM-PDS,effective cluster analysis is expected by calculating distances based on dimensions with common values between vectors containing missingvalues. Additionally, we verify the recommendation performance of the proposed method through numerical experiments using real-worlddatasets.