Using wordnet-based neighborhood for improving social tag recommendation

  • Authors:

  • Ya-Tao Zhu;Sheng-Hua Liu;Xue-Qi Cheng;Yue Liu;Yuan-Zhuo Wang;Jin-Gang Liu

  • Affiliations:

  • Capital Normal University, Beijing, China,Institute of Computing Technology, Chinese Academy of Sciences, China,College of Information Science & Technology, Agricultural University of Hebei, C ...;Institute of Computing Technology, Chinese Academy of Sciences, China;Institute of Computing Technology, Chinese Academy of Sciences, China;Institute of Computing Technology, Chinese Academy of Sciences, China;Institute of Computing Technology, Chinese Academy of Sciences, China;Capital Normal University, Beijing, China

  • Venue:
  • ICIC'12 Proceedings of the 8th international conference on Intelligent Computing Theories and Applications
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recent years have seen social tag recommendation growing into a popular service for users to organize and share digital content on social webpages. Among of knowledge discovery techniques that are applied in social tag recommendation systems, the collaborative filtering based ones are achieving widespread success. The similarity measurement is critical to determine the appropriateness of the results recommendation in the collaborative-filtering schema. In this paper, a nugget is introduced as an atomic conceptual entity generating from WordNet, to measure the similarity of web content and recommend tags. With the nuggets, we can use the WordNet-based neighbors, rather than the literal ones for collaborative filtering, which considers the common sense that the expression varies for a specific concept. The experiments conducted on the dataset from Del.icio.us, have shown that our approach is effective and consistently achieves better precision and recall than both baselines.