International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463 Vol. 2 Issue 6, June-2013, pp: (71-75), Available online at: www.erpublications.com Page | 71 Discovery of frequent subgroup using data mining Seema Mishra Robotics and AI Lab, Indian Institute of Information Technology, Allahabad, India Abstract: Social network analysis is emerging technology that focuses on pattern of interaction/relation among people, organization. Frequent subgroup detection is subpart of social network that reveals the hidden knowledge and reflect the behavior of entire social network. These subgroups are collection of nodes that share common characteristics and densely connected with each other. In this paper, an unexampled approach is acknowledged to discover frequent subgroup inspired from a well known algorithm in the domain of association rule mining recognized as Continuous association rule mining algorithm. Keywords: Subgroup detection, Social network analysis, Dynamic network analysis. I. INTRODUCTION In modern era, social network analysis has been in existence for quite some time and experiencing a surge in popularity to understand the behavior of the users in the form of nodes in the network. In order to model the social network, most popular data structure typically known as graphs are used where the nodes depict the individual or group of person, or event or organization etc and each link/edge represents connection/relationship between two individual [7 ,8]. Social network analysis attempts to understand the network and its components like nodes (social entities commonly known as actor or event) and connections (inter-connection, ties, and links). It has main focus of analyzing individuals and their relationships among them rather than individuals and their attributes as we deal in conventional data structure. Social Network analysis has been in existence from past but now a day’s extensively used to analysis the structure and connection between various actors existing within organization. The ability to detect community structure in a network could have practical applications. Communities in a network might represent real social groupings oftentimes interacting, perhaps by interest or background; communities in a citation network might represent related papers on a single topic; communities in a metabolic network might represent cycles and other functional groupings; communities on the web might represent pages on related topics; hidden communities might represent potential suspicious activity. Being able to identify these communities could help us understand and exploit these networks more effectively. Communities of practice are the collaboration groups that naturally grow and coalesce within any kind of networks. Any institution that provides opportunities for communication or interaction among its members is eventually threaded by communities who have similar goals and a shared understanding of their activities. These communities have been the subject of much research as a way to uncover the structure and interaction patterns within a network in order to understand the collective behavior of the network from the individuals that constitute the network. Recent Research on these networks has focused on using a social network perspective to analyze these networks. A social network consists of both a set of actors, who may be arbitrary entities like persons or organizations, and one or more types of relations between them, such as information exchange or economic relationship. Subgroup detection aims at clustering nodes in a graph into subgroups that share common characteristics. But to some extent, sub graph discovery does the same job for finding interesting or common patterns in a graph. One of the most common interests of social network analysis is the substructures that may be present in the network. Subgroups are subsets of actors among whom there are relatively strong, direct, intense, frequent, or positive ties. From the ideas of subgroups within a network, we can understand social structure and embeddedness of individuals. Finding frequent groups in graph database can be modeled as (a) graph transaction setting and (b) single graph setting. Graph transaction setting takes as input relatively small graph of user interaction whereas single graph setting deals with large graph of user’s interaction involved in communication [M. Kuramochi and G. Karypis, 2004]. The approach we espoused for discovering frequent subgroups is based on continuous association rule mining algorithm [15]. The main aim of adopting this approach is, because the groups of people is not static, it changes over a period of time as the member of group is being joining and leaving from group.