http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Detecting the Hierarchical Community Structure Based on Advanced LPA
Pengtao Jia,Sha Chao,Shuang Qiu 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.8
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.