Diversifying Tag Selection Result for Tag Clouds by Enhancing both Coverage and Dissimilarity

Abstract

Tag cloud has been a popular facility used by social sites for online resource summarization and navigation. Tag selection, which aims to select a limited number of representative tags from a large set of tags, is the core task for creating tag clouds. Diversity of tag selection result is an important factor that affects user satisfaction. Information coverage and item dissimilarity are two major perspectives for exploring the concept of diversity, while existing tag selection approaches usually consider diversification from single perspective. In this paper, we propose a new approach for diversifying tag selection result, which takes into account both information coverage and tag dissimilarity. We design two sub-objective functions about information coverage and tag dissimilarity, respectively, and construct an objective function as a convex combination of the two sub-objective ones. We also give out a greedy algorithm that can well approximate the objective function. We conduct experiments on 17 datasets extracted from the website of CiteULike to compare our approach with existing ones. The experiment results show that our approach can achieve promising performance of diversification.