ABSTRACT
Social data mainly comprise open source social network information from newswire and social media. There are cases, however, where social data can stem from other sources of network data, such as smart phones, proximity sensors, simulated data, surveys, communication networks, private company data, social science research, and databases [CDW13]. Social data exhibit three characteristics that play a key role in social network analysis. First, social data are voluminous. Indeed, there are numerous on-line social networks that allow their users to easily create and share content. For example, there are more than 1.39 billion monthly active Facebook users [FbS15] and around 288 million monthly active Twitter users that generate 5800 tweets per second [Twi15], while 300 hours of video are uploaded to YouTube every minute [You15]. Second, social data are multi-faceted. They range from textual, image, audio, and video content to user metadata. They can be important in many
fields of study besides computer science, such as sociology, politics, or marketing for example. Third, the data are dynamic. Structural changes can occur at multiple time scales or can be localized to a subset of users [LCSX11]. Consequently, social media data analysis needs to handle the data volume as well as the number and the diversity of the data facets in a dynamically evolving framework.
