SocialSearch A Social Platform for Web 2.0 Search Claudio Biancalana, Fabio Gasparetti, Alessandro Micarelli and Giuseppe Sansonetti Department of Engineering, Artificial Intelligence Laboratory, Roma Tre University, Via della Vasca Navale, 79, 00146 Rome, Italy Keywords: Query Expansion, Social Bookmarking Services, Personalization. Abstract: In the last decade, social bookmarking services have gained popularity as a way of annotating and categoriz- ing a variety of different web resources. The idea behind this work is to exploit such services for enhancing traditional query expansion techniques. Specifically, the system we propose relies on three-dimensional co- occurrence matrices, where the further dimension is introduced to represent categories of terms sharing the same semantic property. Such categories, named semantic classes, are related to the folksonomy mined from social bookmarking services such as Delicious, Digg, and StumbleUpon. The paper illustrates a comparative experimental evaluation on real datasets, such as the one collected by the Open Directory Project and the TREC 2004. We also include the results of a specific disambiguation analysis aimed to evaluate the effective- ness of our approach in comparison with state-of-the-art techniques when satisfying queries characterized by polysemic and ambiguous terms. 1 INTRODUCTION The Social Semantic Web combines together the core principles of the Semantic and Social Web: it in- cludes, on the one hand, the idea of associating a se- mantic description with web resources for enabling machines to access and process them, on the other hand, the idea of exploiting social content information for that purpose. This development, however, leads to the need to revise the classical techniques for the tra- ditional Web (Micarelli et al., 2006; Lops et al., 2007; Gentili et al., 2001; Gasparetti and Micarelli, 2003), as they could not be more efficient in the new Web design. Automatic query expansion (QE) is a well-known technique for enabling users to better characterize their search domain by supplementing the original query with additional terms that are somehow linked to the frequency of the term the user specified in his query (Bai et al., 2005). This method can significantly improve the performance of information retrieval sys- tems. However, traditional QE techniques, even those providing users with personalized results, may suffer from some drawbacks if extended to the Social Se- mantic Web. In particular, additional terms can (i) be simple synonyms, (ii) not consider the existence of different lexicons, given that each user has his own custom dictionary. As a result, QE process may fail to contextualize the research domain of interest if multi- ple users annotate the web content. Our research objectives include (i) to find a solu- tion to the lack of expression of the candidate terms for query expansion, (ii) to customize the search re- sults taking into account the semantic domain of the user interests, (iii) more generally, to explore novel approaches combining semantic, social, and adaptive aspects. The proposed system - named SocialSearch - is an extension of the traditional QE techniques, which are based on the computation of two-dimensional co- occurrence matrices (Biancalana and Micarelli, 2009; Biancalana et al., 2012). Our approach makes use of three-dimensional co-occurrence matrices, where the added dimension is represented by semantic classes (i.e., categories comprising all the terms that share a semantic property) related to the folksonomy ex- tracted from social bookmarking services (Musto et al., 2009) such as Delicious 1 , Digg 2 and Stum- bleUpon 3 . These web sites allow users to store, or- ganize, share, and search bookmarks associated with web resources, through the input of additional data 1 delicious.com 2 digg.com 3 www.stumbleupon.com 70 Biancalana C., Gasparetti F., Micarelli A. and Sansonetti G.. SocialSearch - A Social Platform for Web 2.0 Search. DOI: 10.5220/0004943000700081 In Proceedings of the 10th International Conference on Web Information Systems and Technologies (WEBIST-2014), pages 70-81 ISBN: 978-989-758-023-9 Copyright c  2014 SCITEPRESS (Science and Technology Publications, Lda.)