Abstract | ||
---|---|---|
It's important to detect the community structure in a network. We can use the information to understand the characteristics of the network and develop applications based on that. A popular approach finds out dense sub graphs from a graph converted from the network, and each resulting subgraph is regarded as a distinct community. However, not every node in the graph must belong to some community , causing a big challenge to this approach. In this paper, we propose a method to detect dense subgraphs in an undirected and unweighted graph with the adoption of punished similarity. The similarity between a pair of nodes in the graph is multiplied by the ratio of the length of the shortest path between the nodes to the diameter of the graph. Experimental results show that our proposed method can achieve better performance than other methods. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/ICMLC.2013.6890751 | ICMLC |
Keywords | Field | DocType |
punished similarity,dense sub graph,shortest path,dense subgraphs,graph theory,network community structure detection,hierarchical clustering,network theory (graphs) | Block graph,Graph,Discrete mathematics,Community structure,Shortest path problem,Computer science,Distance,Artificial intelligence,Clique-width,Longest path problem,Machine learning | Conference |
Volume | ISSN | Citations |
03 | 2160-133X | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rong-Fang Xu | 1 | 14 | 1.23 |
Hung-Wen Peng | 2 | 0 | 0.34 |
Shie-Jue Lee | 3 | 48 | 5.11 |