A Novel Approach of Discovering Local Community Using Node Vector Model
| Autor: | Daling Wang, Jinglian Liu, Yifei Zhang, Weiji Zhao, Shi Feng |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Current (mathematics)
Jaccard index Computer science Node (networking) Community structure Network structure 020207 software engineering 02 engineering and technology computer.software_genre Local community Global information 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Data mining computer |
| Zdroj: | Web Information Systems Engineering – WISE 2016 ISBN: 9783319487397 WISE (1) |
| DOI: | 10.1007/978-3-319-48740-3_38 |
| Popis: | Local community detection aims at discovering a community from a seed node without global information about the entire network structure, and various local community detection algorithms have been proposed. However, most existing algorithms either are parameter-dependent or have low accuracy. In this paper, we propose a novel approach of discovering local community using node vector model. In detail, we propose node vector model to represent nodes in graphs. Moreover, we define weighted Jaccard similarity coefficient to estimate the similarities between nodes. Based on the model and definition, local community can be detected. Our algorithm gives priority to the node which is most similar to the nodes in the current local community. We compare the proposed algorithm on both synthetic and real-world networks. The experimental results demonstrate that our algorithm is highly effective at local community detection compared to related algorithms. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|