In this chapter, we introduce a technical foundation for mining social data. We do so through a set of examples that focus on Twitter data, though the analyses are equally appropriate for data from other venues. That being said, our use of Twitter warrants a bit more explanation: what's so interesting about Twitter?

One answer to this question is nothing. Twitter is one of several social media networks, and there is little reason to suspect that data from Twitter is fundamentally different from other socially generated data. Another answer is that Twitter is different in subtle but important ways. One distinction is Twitter's ability to foster second-order connections, or what Granovetter (1973-1983) calls weak ties. These weak ties are important as they bring information to individuals from those with whom they share less, thus dramatically increasing information exposure. Second, Twitter, perhaps more than some other social networks, allows users to self-organize. A third answer is that Twitter users actively use Twitter to gather insight, make recommendations, and lodge public complaints. The extent to which users find this information valuable gives credence to the notion of its validity.