A method for personalized ranking of items based on similarity between Twitter users
Taketoshi Ushiama, Kazushige Tominaga · 2014
According to the remarkable progress of Web technologies, people search various types of items such as movies, books, and CDs on the Internet. Items that are returned as search results would be ranked based on objective criteria such as rating scores. Such rankings are not personalized to a user based on his/her interests and preferences. In this paper, we introduce a method for ranking items that are specified by users. Our method does not request users to specify their interests and preference manually. Our method focuses on the users who have posted tweets about the items and extracts features of the users. Items are ranked according to the preferences of the user based on the similarities of twitter users. The effectiveness of the proposed method is evaluated with experimental results on subjects.