1C3-2 ファジィc平均法を用いた協調フィルタリングにおける一考察(ソフトコンピューティング)

概要

論文の詳細を見る
Collaborative filtering is a useful technique for reducing information overload in network communities and is regarded as a task for connecting promising items to users with the goal being to recommend the items. In neighborhood-based approaches, it was shown that the recommendation ability can be improved by several strategies. In this paper, Fuzzy c-Means clustering algorithm is used for extracting user-neighborhood and the weighted average is calculated by considering fuzzy membership degrees of users. In comparative experiments, the several strategies used in neighborhood-based approaches are applied and discussed the applicability of them.
一般社団法人日本機械学会の論文
2011-09-01