Social Community-Oriented Social Attribute Analysis: An Empirical Study on QQ Group Data

Social networks have got increasing attention in recent years, and Social Networking Services (SNS), such as QQ, Facebook and Twitter, are among the most popular network applications on the internet. The popularity of these SNS attract more and more researchers to focus on the study of attributes of social network, especially the groups which are ubiquitous in social networks, both to improve current systems and to design new applications of social networks. In this paper, we study statistically 1.4 billion personal social information from QQ and 80 million group social information from QQ, and systemically analysis the distribution of attributes of all these social context data. And then we discover that both the frequency of age and the frequency of age for each gender conform to normal distributions. In addition, we analysis the QQ user registration time and the relationship between digital number of QQ and their age. Moreover, the distribution of the number of users in QQ group conforms to a power-law distribution. Furthermore, as the KL distance between the sample data and the whole data is within 0.1, i.e. the similarity of the sample data and the whole data is high, we can use the results based on the sample data to estimate the properties of the whole data in the integrated QQ network.