Pattern Mining on Ego-Centric Networks of Friendship Networks
The paper proposes a procedure to analyse local patterns of connectivity among people in social networks using the idea of ego-centric network. The ego-centric networks of every nodes are transformed into normalized representation and classified into patterns. The procedure can be applied to large dataset by giving it in SQL code. We applied the procedure to friendship networks and demonstrated distinguished properties compared to other networks. We found out that friendship network contains large variety of patterns.
KeywordsSocial Network Complete Graph Average Degree Random Network Frequent Pattern
Unable to display preview. Download preview PDF.
- 2.Inuzuka, N., Nakano, T., Shimomura, K.: Friendship Analysis Using Attendance Records to University Lecture Classes. In: Proc. IASK Int’l Conf. Teaching and Learning, pp. 478–486 (2008)Google Scholar
- 3.Klemm, K., Eguiluz, V.M.: Highly clustered scale-free networks. Physical Review E 65, 036123 (2002)Google Scholar
- 4.Matsushima, H., Kadosaka, S., Yamamoto, S., Inuzuka, N.: Analysis of Friendship Network Using Attendance Records to Lecture Classes. In: 30th Sunbelt Conf. (Tech. Rep., Inuzuka labo.) (2010)Google Scholar
- 6.Newman, M.E.J., Girvan, M.: Finding and evaluating community structure in networks. Physical Review E 69, 026113 (2004)Google Scholar