Abstract
Currently many factors can influence community detection in mobile social networks, where node mobility is a key factor to influence the stability of community structure. In this paper, we propose a social community detection scheme for mobile social networks based on social-ware, including social attribute similarity, node interest similarity and node mobility. Compared with other community detection schemes, our proposed scheme can accurately detect the communities based on community attribute and node mobility. The experiments show the numbers of detected communities and members in the maximal size community generated by our scheme are both smaller than those of the GN and NM schemes. Additionally, since the nodes (users) both have higher mobility in mobile social networks, the efficiency of our proposed scheme relatively becomes higher. The experiments show when the values of mobility in the test data sets increase, the running time of our proposed scheme decreases when the number of edges is fixed. For example, the running time of our proposed scheme is about 17s when the maximum value of mobility is set as 0.5 and the number of edges is about 16000, and further the running time is only about 13s when the maximum value of mobility is set as 1. Therefore, our proposed scheme can more accurately and efficiently make community detection to increase the stability of mobile community structure.
Highlights
INTRODUCTIONMany community detection schemes [14]–[23] were proposed to divide social network structure: 1) graph-based partitioning algorithm [14]; 2) module degree algorithm [15]; 3) edge clustering algorithm [16], [17]; 4) hierarchical clustering algorithm [18]; 5) seed dispersal method [19], [20]; 6) random walk algorithm [21]; 7) label propagation method [22], [23]
Our proposed scheme can more accurately and efficiently make community detection to increase the stability of mobile community structure
OUR CONTRIBUTIONS In this paper, we propose a social community detection scheme for mobile social networks based on social-ware, including social attribute similarity, node interest similarity and node mobility
Summary
Many community detection schemes [14]–[23] were proposed to divide social network structure: 1) graph-based partitioning algorithm [14]; 2) module degree algorithm [15]; 3) edge clustering algorithm [16], [17]; 4) hierarchical clustering algorithm [18]; 5) seed dispersal method [19], [20]; 6) random walk algorithm [21]; 7) label propagation method [22], [23]. Based on the change of topology of mobile social networks, many researchers proposed some schemes to detect community structure. Guan and Wu [53] proposed a novel method that effectively divides social communities according to the human activity characteristics in opportunistic networks [54] Their established effective information transmission based on structuralization areas scheme allows information to be transmitted between resource nodes and communities [55]. The process of community detection and classification is based on the modularity maximization principles: 1) the terms of negative value should be excluded; 2) the nodes with greater similarity should be partitioned to the same communities
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
