期刊名称：International Journal of Multimedia and Ubiquitous Engineering
印刷版ISSN：1975-0080
出版年度：2015
卷号：10
期号：8
页码：207-214
DOI：10.14257/ijmue.2015.10.8.21
出版社：SERSC
摘要：LPA is a classical community detection algorithm with linear time complexity. It can be applied to networks whose community structure is known. However, the randomness of it's initial nodes causes a poor stability. In this paper, we proposed a new method to improve the algorithm by reordering the initial nodes. The advanced algorithm uses ascending order instead of descending order based on degree. It can improve the stability of the original algorithm and detect the hierarchical community structure of the complex network. We tested the algorithm on Zachary's karate network, Dolphins social network and American football network. The experiments confirmed that compared to the classical algorithm the improved algorithm has better performance in terms of rationality and accuracy.
