For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Extracting User Interest for User Recommendation Based on Folksonomy
Junki SAITO Takashi YUKAWA
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2011/06/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Data Engineering, Web Information Systems
user profiling, folksonomy, social bookmarking, SNS, Twitter,
Full Text: PDF(232.5KB)>>
In the present paper, a method for extracting user interest by constructing a hierarchy of words from social bookmarking (SBM) tags and emphasizing nouns based on the hierarchical structure (folksonomy) is proposed. Co-occurrence of the SBM tags basically have a semantic relationship. As a result of an experimental evaluation using the user profiles on Twitter, the authors discovered that the SBM tags and their word hierarchy have a rich vocabulary for extracting user interest.