Group member’s importance evaluation algorithm based on identity similarity

Abstract:
Research on the importance ranking of members in a certain group is an important branch in social-network, but the existing importance algorithms ignore the influence of the members' identities differences on the importance assessment. To solve this problem, this paper proposes a K-shell Level based on Identity Similarity (KSL-IS) algorithm to measure the importance of social network members. According to the group members’ link structure, the KSL-IS algorithm defines the identity similarity among members, integrates the location characteristics of members and the influence of neighbourhood members, and comprehensively evaluates the importance of members. This study used four different complex networks as experimental data, compared with existing algorithms, and verified by the Susceptible-Infectious (SI) propagation performance experiment, robustness test, and Kendall correlation coefficient. The investigation showed that the KSL-IS algorithm is more accurate to measure the importance of members in the group. This study verifies that identity and location characteristics should be taken into account when calculating the importance of members.