Social network analysis: Evolving Twitter mining

The growth of techniques of social network analysis is fast at present. These techniques are of interest to many researchers in different areas such as sociology, communication and computer science, social psychologist and so on. Nowadays, by analyzing how the members of network interact, share information or establish relationships, useful knowledge about them and their relations can be extracted. However, information related to how these members are presented to the world (by their users profiles) could give use also very useful knowledge. In this paper, we present an approach to automatically analyze the Twitter user profiles of a specific community of users. The locations of these users can also be selected by the user. The proposed analysis is done by extracting some characteristics of the collected profiles (of that given community). This analysis includes the detection of outliers, the clustering of profiles and their classification. The most important characteristic of the presented approach is that it can cope with the data of thousands of twitter profiles in real-time. Thus, this work is related to big data in the area of big data analytics. The approach presented in this paper is based on evolving fuzzy systems, which makes possible not only that we can cope with thousands of data in real-time, but also that the knowledge that we obtain from the social networks can be constantly updated (evolving).