Analyzing Social Networks to Mine Important Friends

Analyzing Social Networks to Mine Important Friends

Carson K.-S. Leung (University of Manitoba, Canada), Irish J. M. Medina (University of Manitoba, Canada) and Syed K. Tanbeer (University of Manitoba, Canada)
Copyright: © 2013 |Pages: 15
DOI: 10.4018/978-1-4666-2806-9.ch006
OnDemand PDF Download:
No Current Special Offers


The emergence of Web-based communities and social networking sites has led to a vast volume of social media data, embedded in which are rich sets of meaningful knowledge about the social networks. Social media mining and social network analysis help to find a systematic method or process for examining social networks and for identifying, extracting, representing, and exploiting meaningful knowledge—such as interdependency relationships among social entities in the networks—from the social media. This chapter presents a system for analyzing the social networks to mine important groups of friends in the networks. Such a system uses a tree-based mining approach to discover important friend groups of each social entity and to discover friend groups that are important to social entities in the entire social network.
Chapter Preview


Due to advances in technology, Web-based communities and social networking sites have emerged. They have facilitated collaboration and information sharing between users, and have led to a vast volume of social media data. Intuitively, social media can be considered as forms of electronic communication (e.g., Web sites such as Twitter for social networking and micro-blogging) through which users create online Web-based communities to share information, ideas, personal messages, and other contents such as images and videos. More formally, social media (Kaplan & Haenlein, 2010) refer to groups of electronic communications (e.g., for Web-based or mobile-based applications [Matera, 2009]) that (a) build on the ideological and technological foundations of Web 2.0 and (b) allow the creation and exchange of user-generated content. Common social media include (a) blogs or tweets (e.g., Blogger, Twitter [Kelsey, 2010]), (b) collaborative projects (e.g., Wikipedia), (c) content communities (e.g., Flickr, SlideShare, YouTube [Lacy, 2008]), (d) social networking sites (e.g., Facebook, Google+, LinkedIn, MySpace), (e) virtual game worlds (e.g., multiplayer online role-playing games like EverQuest and its successors, as well as World of Warcraft), and (f) virtual social worlds (e.g., Second Life [Rymaszewski, et al., 2007]). Embedded in these social media data are rich sets of meaningful knowledge. This leads to social media mining, which aims to extract, represent, and exploit the rich sets of meaningful knowledge from vast volumes of social media data ranging from those in digital textual forms to those in rich multimedia formats.

Social media mining can also be considered as an interaction between data mining and social computing. Data mining (Frawley, et al., 1991) refers to non-trivial extraction of implicit, previously unknown, and potentially useful information from data (e.g., social media data); social computing intersects social behaviour and computing systems in the sense that it computationally facilitates social studies and human-social dynamics in social networks, creates social conventions through the use of computer software, and designs information and communication technologies to deal with social context. An important social media-mining task is to discover meaningful knowledge about the social networks residing in the social media data. A social network (Schwagereit & Staab, 2009) is a structure made of social entities (e.g., individuals, corporations, collective social units, or organizations) that are linked by some specific kinds of interdependency (e.g., friendship, kinship, common interest, beliefs, or financial exchange). A social entity is connected to another entity as his friend, next-of-kind, collaborator, co-author, classmate, co-worker, team member, or business partner. As such, social network analysis (Wasserman & Faust, 1994)—which often refers to a systematic method or process for identifying useful social information like interdependency among social entities (i.e., social relationships and connections) from the social networks—is in demand.

Nowadays, various social computing applications such as blogs, email, instant messaging, social bookmarking, social networking, and wikis have been widely popularized so that people could interact socially via computing space. For instance, a Facebook user can create a personal profile, add other users as friends (who can be further categorized into different customized lists such as close friends, acquaintances, or family), exchange private messages, post messages on friends’ walls, and join common-interest user groups. Similarly, a LinkedIn user can create a professional profile, establish connections to other users (who can be further annotated with tags corresponding to overlapping categories such as colleagues, classmates, business partners, and friends), exchange messages, recommend other users, and join common-interest user groups. A Google+ user can create a profile and add other users in one or more of his circles (e.g., circles for his friends, family, acquaintances, and followers). He can also share some posts, photos, or videos among users in the same circles.

Complete Chapter List

Search this Book: