Detection of Opinion Leaders in Social Networks Using Entropy Weight Method for Multi-attribute Analysis

Abstract

Opinion leaders are those users who have great influence, whose emergence has an important effect on social networks. Thus, the recognition of opinion leaders can contribute to a comprehensive understanding of the development trend of information dissemination and other applications. This paper proposes a new opinion leaders detection algorithm, called MAA algorithm, which uses Entropy Weight Method to comprehensively multiple attributes analysis including location attributes, distance attributes, and strength entropy. Specifically, our main contributions are the following two aspects: (1) Multiple attributes analysis is conducted to measure the influence of a user in the dissemination of information; (2) Entropy Weight Method is used to comprehensively evaluate multiple attributes, and calculate the influence weight, respectively. Experimental results illustrate that the proposed algorithm is superior to other traditional algorithms.