Tracking dynamic community evolution in social networks

Healthcare social networking has known a prominent popularity. Nowadays, millions of healthcare communities are appeared. Community detection methods in social networks have essentially focused on static graphs neglecting the temporal characteristics of networks. Recently, motivated by the dynamic nature of real world projected on virtual social networks, an increasing number of evolutionary properties of communities issues are considered by research community. In this work, we aim to present a study on social networks across the time axis i.e. temporal social networks and we propose an algorithm classifying changes and based on indicators, and also we integrate data warehouse layer in order to have an over-view of all possible changes helpful for future analysis. The experimental results include an example of changes related to the apparition of diabetes in our country Tunisia.