The Strength of Considering Tie Strength in Social Interest Profiling

Abstract

With the emergence of social networking platforms and great amount of generated content, analyzing people interactions and behaviour raises new opportunities for several applications such as user interest profiling. In this context, this paper highlights the importance of considering relationship strength to infer more refined and relevant interests from user’s direct neighbourhood. We propose WeiCoBSP, a Weight-aware Community-Based Social Profiling approach that leverages strength of ego-friend and friend-friend relationships. The former, describing connections with the profiled user, allows to identify most relevant people from whom to infer worthwhile interests. The latter qualifies connections among user’s neighbourhood and enables depicting the most realistic community structure of the network. We present an empirical evaluation performed on real world co-authorship networks, validating our approach. Experimental results demonstrate the ability of WeiCoBSP to infer user’s interest accurately, improving greatly the unweighted CoBSP process but also results of experiments assessing separately ego-friend and friend-friend relationships strength.