The innovative technologies and advanced services typical of the so called Web 2.0 have led to the development of many online social networking Web sites where users can collaborate and share information, interests, personal and professional activities. These sites integrate new technologies and services, which allow users to participate to the Web not only as consumers of the contents centrally uploaded by the providers, but also as producers of contents and consumers of contents provided by other users. Hence, the workload of these sites is the result of the complex social relationships that can be developed among users.

This research project is aimed at studying and characterizing the workload of social networking sites in terms of the characteristics and behavior of the users and the communities of users sharing common interests. The objective is to derive models that describe the user behavior and the interactions between the users and the services offered within the sites.

In particular, the Big Data stored by online platforms for emotional support offering free, anonymous, and confidential conversations with live listeners are considered. This study explores the utilization and the interaction features of hundreds of thousands of users. It dissects the user's activity levels, the patterns by which they engage in conversation with each other, and uses machine learning methods to find factors promoting engagement. Graph analysis techniques are applied to describe community structure and evolution, and characteristics such as the emergence of a rich-get-richer phenomenon in the development of the network.