Folksonomy Reasoning based Personalized Tagging and Tag Recommendation for Web Page Search

Abstract

Provision of high-quality tags for huge amounts of untagged web resources has recently become a hot topic. Folksonomy, a famous classification technology, reflects users' behaviors on web page searching and users' interests in web pages. However, folksonomy has some drawbacks. The main drawback is the lack of semantic information. We propose a technique using folksonomy reasoning to resolve this issue. The purpose of folksonomy reasoning is to eliminate ambiguity of tags by semantic concepts analysis. The technique helps users tag web pages and recommend related tags to users. The accuracy of personalized tagging has been improved by applying folksonomy reasoning. The user feedback mechanism allows users to rate recommended tags in our technique. We demonstrate the feasibility of personalized tagging based on folksonomy reasoning and tag recommendation in web page search using a case study with our prototype system.