International Journal of Web & Semantic Technology (IJWesT) Vol.4, No.2, April 2013 DOI : 10.5121/ijwest.2013.4202 9 APPLICATION OF CLUSTERING TO ANALYZE ACADEMIC SOCIAL NETWORKS K. Sobha Rani 1 , KVSVN Raju 2 and V.Valli Kumari 3 1 Dept. of Information Technology, MVGR College of Engineering, Vizianagaram. sobharani.t@gmail.com 2 Anil Neerukonda Inst. of Technology & Sciences, Visakhapatnam. kvsvnraju@gmail.com 3 Department of CSSE, College of Engineering, Andhra University, Visakhapatnam. vallikumari@gmail.com ABSTRACT Social network is a group of individuals with diverse social interactions amongst them. The network is of large scale and distributed due to involvement of more people from different parts of the globe. Quantitative analysis of networks is need of the hour due to its’ rippling influence on the network dynamics and in turn the society. Clustering helps us to group people with similar characteristics to analyze the dense social networks. We have considered similarity measures for statistical analysis of social network. When a social network is represented as a graph with members as nodes and their relation as edges, graph mining would be suitable for statistical analysis. We have chosen academic social networks and clustered nodes to simplify network analysis. The ontology of research interests is considered to measure similarity between unstructured data elements extracted from profile pages of members of an academic social network. KEYWORDS: Social Network Analysis, Clustering, Graph Mining, RDF 41. INTRODUCTION Different kinds of network exist viz. social, technological, business etc., all of which share similar attributes like being distributed, continuously growing and of large scale. There are some interesting quantifiable measures that help analyze these networks, like number of nodes, connectivity, centrality, clustering coefficient and degree distribution [14]. These networks can be modeled as random graphs, scale-free networks and hierarchical networks. Due to the property of being scale free, the social networks continuously expand with addition of new nodes and the relation amongst the nodes vary with time and frequency of interaction between them. Also the probability of a node being influential in the network is not uniform due to their difference in attributes and interactions. The social network data is not even suitable to be stored in relational database; hence different approaches are followed to store and analyze unstructured social network data. 1.1 ACADEMIC SOCIAL NETWORKS With the growth of the internet and the World Wide Web, social networks have become influential. One of the most effective channels for obtaining information is the informal network of collaboration of colleagues and friends etc. The use of social networks is widespread in employers’ recruiting and in workers’ job-seeking, pursuit of hobbies and building collaboration within or between organizations. A person can have different types of information: personal profile with fields like homepage, field of interest and hobbies, contact information including