A semantic-based approach for searching and browsing tag spaces

In this thesis we propose the Semantic Tag Clustering Search framework (STCS). This framework consists of three parts. The first part deals with syntactic variations by clustering tags that are syntactic variations of each other and assigning a label to them. The second part of the framework addresses the problem of recognizing homonyms and identifying semantically related tags. The last, and final part of the STCS framework utilizes the clusters obtained from the first two parts to improve search and exploration in tag spaces. For removing syntactic variations, we use the normalized Levenshtein distance, and the cosine similarity measure based on tag co-occurrences. For creating semantic clusters, we employ two non-hierarchical, and two hierarchical clustering techniques. To evaluate the value of the semantic clusters, we develop an Web Application called XploreFlickr.com.