Community Detection and Profiling in Location-Based Social Networks

Community Detection and Profiling in Location-Based Social Networks

Zhu Wang (Northwestern Polytechnical University, China), Xingshe Zhou (Northwestern Polytechnical University, China), Daqing Zhang (Institut Telecom SudParis, France), Bin Guo (Northwestern Polytechnical University, China) and Zhiwen Yu (Northwestern Polytechnical University, China)
DOI: 10.4018/978-1-4666-4695-7.ch007
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Due to the proliferation of GPS-enabled smartphones, Location-Based Social Networking (LBSNs) services have been experiencing a remarkable growth over the last few years. Compared with traditional online social networks, a significant feature of LBSNs is the coexistence of both online and offline social interactions, providing a large-scale heterogeneous social network that is able to facilitate lots of academic studies. One possible study is to leverage both online and offline social ties for the recognition and profiling of community structures. In this chapter, the authors attempt to summarize some recent progress in the community detection problem based on LBSNs. In particular, starting with an empirical analysis on the characters of the LBSN data set, the authors present three different community detection approaches, namely, link-based community detection, content-based community detection, and hybrid community detection based on both links and contents. Meanwhile, they also address the community profiling problem, which is very useful in real-world applications.
Chapter Preview
Top

1. Introduction

The recent surge of location-based social networks (LBSNs, e.g., Foursquare and Facebook Places) driven by the increasing popularity of smart phones is bringing a new set of opportunities for research scientists and application developers. Compared with traditional online social networks (e.g., Facebook, Twitter), a distinct characteristic of LBSNs is the co-existence of both online and offline social interactions, as shown in Figure 1. On one hand, LBSNs support typical online social networking facilities, e.g., making friends, sharing comments and photos. On the other hand, LBSNs also support offline social interactions, e.g., checking in places. In other words, LBSNs are heterogeneous social networks which consist of both online and offline social links (Guo, Zhang, Wang, Yu & Zhou, 2013). Meanwhile, vertices in LBSNs usually have multiple attributes, e.g., attributes of a user might include number of followers, number of followings, and number of check-ins; a venue might have attributes such as category, number of check-ins and number of visitors.

Figure 1.

An example of LBSN

One fundamental issue in social network analysis is to detect user communities. A community is typically thought of as a group of users who are densely interconnected compared to the other users in the network (Newman & Girvan, 2004; Fortunato, 2010). Specifically, discovering communities of LBSN users who visit similar physical places are able to facilitate many applications, such as direct marketing, friend recommendation, and community sensing (Zhang, Guo & Yu, 2011). However, unlike social networks (e.g., Flickr, Facebook) which provide explicit groups for users to subscribe or join, the notion of community in LBSNs is not well defined. In order to capitalize on the huge number of potential users, quality community detection and profiling approaches are needed.

Firstly, unlike social networks which only contain a single type of social interaction, the co-existence of online/offline social interactions and user/venue attributes in LBSN makes the community detection problem much more challenging.

Meanwhile, it has been well understood that people in social networks are naturally characterized by multiple community memberships. For example, a person usually belongs to several social groups like family, friends and colleagues; a researcher may be active in several areas. Thus, it is more reasonable to cluster users into overlapping communities rather than disjoint ones.

Finally, we believe that it’s important to characterize communities in a semantic manner to effectively support real-world applications. However, due to the limitation of available information, not much work has been done to address community profiling. The rich user and venue metadata available in LBSNs provides us the possibility to characterize the identified communities semantically.

In this chapter, we will present some of the recent progress in community detection and profiling on LBSNs. Based on the used information for LBSN community discovering, we classify existing community detection approaches into three categories:

  • Link-based community detection in LBSNs, which mainly leverages the online/offline social links among users to discover communities.

  • Content-based community detection in LBSNs, which performs community clustering mainly using the attributes of users.

  • Hybrid community detection based on both links and contents in LBSNs, which explores both the online/offline social links and the user/venue attributes to detect social communities.

The rest of this chapter is organized as follows. In Section 2, we analyse the characteristic of the user’s digital footprints in LBSNs. The above mentioned three different community detection approaches are presented in Section 3, 4 and 5, respectively. Finally, we conclude this chapter by proposing some promising research directions.

Complete Chapter List

Search this Book:
Reset