A New Model for Discovering Communities of Users on Social Network

Abstract

The trend of technological development and increasing varieties of social media lead to the changes in people’s behaviors in society and forming online communities. Changes of human’s behaviors make many models of business, marketing, services and even the field of education, security, politicsl change from approaches to user management. Community of users on social networks influence behaviors, habits of each user involved in the community. Therefore, exploring community on social networks from many different data sources via analyzing exchanged contents will help know the user community’s behaviors which are reflected in the content and topics that users are interested in discussing in messages. In this paper, we propose a new model of discovering communities of users on social networks based on the topic model combined with Kohonen network. In the proposed model, we focus on discovering communities of users on social networks and analyzing the interested topics change of online community in each period of time. The proposed model is experimented with a set of vectors in interested topics of online users in higher education field.