This chapter discusses how data mining techniques differ from traditional quantitative methods and illustrates their current applications in the social sciences. First it introduces the definition and scope of data mining research as well as its epistemological premises, resources available, procedures, and potential outcomes. Then it classifies data mining into three groups of studies based on the types of corpus in question: Studies of mainstream media, studies of user-generated content, and studies of meta-data. These groups represent three important dimensions respectively: the growing volume of information, the structural patterns of communication and interaction, and the availability of ‘second-hand’ data or meta-data aggregated by other automatic means. The chapter concludes by presenting the advantages and disadvantages of data mining compared to traditional quantitative methods of data collection, and outlining future directions for data mining research.