A semantic clustering-based approach for searching and browsing tag spaces

Abstract

Many of the existing cloud tagging systems are unable to cope with the syntactic and semantic tag variations during user search and browse activities. As a solution to this problem, in this paper, we propose the Semantic Tag Clustering Search, a framework able to cope with these needs. The framework consists of three parts: removing syntactic variations, creating semantic clusters, and utilizing the obtained clusters to improve search and exploration of 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 improve an existing non-hierarchical clustering technique. Using our framework, we are able to find more clusters and achieve a higher precision than the original method. The advantages of a cluster-based approach for searching and browsing through tag spaces have been exploited in XploreFlickr.com, the implementation of our framework.

A Literature Survey on Ranking Tagged Web Documents in Social Bookmarking Systems

“…To calculate user's interest for personalization as defined by the second parameter is given in equation (3). I u, r = uc_w(u, c) * c∈C rc_w(r, c) (3) where uc w u,c = Numer of times r is annotated with a tag from a cluster c total number of annotations by user u rc w r,c = Number of times the resource r is annotated with a tag form cluster c total number of times the resource r is annotated 3.5.5 Semantic Tag Clustering Search D. Vandic et al [64] proposed the idea of Semantic Tag Clustering Search (STCS) for sorting web documents which is based [65], which is further based on [66]. The sorting formula based on cosine similarity by using equation (1).…”

Semantic Disambiguation and Contextualisation of Social Tags

“…Other works are based on the concept of tag co-occurrence, and aim at extracting tag semantic meanings and contexts within a particular folksonomy by applying probabilistic models and clustering techniques on the tag space according to the tag co-occurrences in item annotation profiles [24,23,18,2,22,9]. For example, for the tag sf, often co-occurring tags such as sanfrancisco, california and bayarea may be used to define the context "San Francisco, the Californian city", while co-occurring tags like sciencefiction, scifi and fiction may be used to define the context "Science Fiction, the literary genre.…”

“…Other works are based on the concept of tag co-occurrence, that is, on extracting the actual meaning of a tag by analysing the occurrence of the tag with others in describing different resources. These approaches usually involve the application of probabilistic models and clustering techniques over the co-occurrence information gathered from a folksonomy [24,23,18,2,22,9], and have been exploited by recent personalisation and recommendation approaches [16,18,17]. Their main advantage is that an external knowledge source is not required.…”

“…However, it has a much higher computational cost since it is applied to large semantic networks. Finally, Vandic et al [22] apply a nonhierarchical strategy for creating semantic clusters. Before the clustering process, their approach removes syntactic variations of tags by using the normalised Levenshtein distance, and the cosine similarity measure based on tag co-occurrences.…”

Reorganizing clouds: A study on tag clustering and evaluation

“…In Vandic et al (2011), the authors propose a method to improve search on social tagging systems by clustering syntactic variations of tags with the same meaning. They use the cosine similarity based on co-occurrence vectors for measuring semantic relatedness.…”