Identify Implicit Social Network by RST/FL Framework

The rapid growth of Internet has transformed the way relationships between companies and their customers. To find the implicit social network and form customer club (community) for customers is a critical requirement in some businesses. The social role analysis attempts to find explicit similarities between actors in the network. Traditional clustering methods based on attributes between actors lack for logic foundation. In this paper, we apply rough set theory to partition objects into equivalence classes interpreting the hidden community. This paper proposes a framework to find the implicit social network based on rough set theory and frame logic to extract and express the social structure and relationship. The development framework includes three levels, i.e. conceptual level, logical level and validation level. In conceptual level, we explore the properties of social network. The logic and validation is based on frame logic. The interface of different level is a mapping from a source model to a target model using a set of transformation rules. Finally, we apply FLORID tools to evaluate the correctness and the adequacy of the model.