Text data from social media and networks are ubiquitous and are emerging at a high rate. Tackling these bulky text data has become a challenging task and an important field of research. The mining of text data and examining it with the help of several clustering techniques, classification techniques, and soft computing methods has been studied in a comprehensive manner in the past. This chapter focuses mainly on the hybrid techniques that have been used to mine textual data from social networks and media data. Social networks are considered a profuse source of viewpoints and outlooks of the public on a worldwide scale. An enormous amount of social media data is produced on a regular basis, generated because of the communication between the users who have signed up for the various social media platforms on several topics such as books, movies, politics, products, etc. The users vary in terms of factors such as viewpoints, scenarios, geographical situations, and many other settings. If mined efficiently, the data have the potential to provide a helpful outcome of an exegesis of social quirks and traits. This chapter offers a detailed methodology on how data mining, especially text mining, is applied to social networks in general. Furthermore, it goes on to introduce the traditional models used in mining the various hybrid methodologies that have evolved and make a comparative analysis. We also aim to provide the future scope and research studies present in this field with all possible new innovations and their applications in the real world. © Springer International Publishing AG 2017. All rights reserved.