Top-r Influential Community Search in Bipartite Graphs
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915495804928000 |
|---|---|
| author | Zhang, Yanxin Hua, Zhengyu Yuan, Long Chen, Zi |
| author_facet | Zhang, Yanxin Hua, Zhengyu Yuan, Long Chen, Zi |
| contents | Community search on bipartite graphs, especially influential community detection, has received significant attention. Existing studies use minimum vertex weights, inadequately reflecting true community influence when some vertices have low weights. To address this, we introduce the $(α,β)$-influential community model based on the average vertex weights from both layers, providing a more comprehensive influence measure. Given the NP-hardness of accurately identifying top-$r$ communities, we propose an exact recursive algorithm enhanced by a slim tree structure and upper-bound techniques to improve efficiency. Additionally, we develop a greedy approximate algorithm with $O((n+m)+m\log{n})$ complexity, further optimized by a pruning strategy. Experiments on 10 real-world graphs demonstrate the effectiveness and efficiency of our proposed algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_06216 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Top-r Influential Community Search in Bipartite Graphs Zhang, Yanxin Hua, Zhengyu Yuan, Long Chen, Zi Social and Information Networks Community search on bipartite graphs, especially influential community detection, has received significant attention. Existing studies use minimum vertex weights, inadequately reflecting true community influence when some vertices have low weights. To address this, we introduce the $(α,β)$-influential community model based on the average vertex weights from both layers, providing a more comprehensive influence measure. Given the NP-hardness of accurately identifying top-$r$ communities, we propose an exact recursive algorithm enhanced by a slim tree structure and upper-bound techniques to improve efficiency. Additionally, we develop a greedy approximate algorithm with $O((n+m)+m\log{n})$ complexity, further optimized by a pruning strategy. Experiments on 10 real-world graphs demonstrate the effectiveness and efficiency of our proposed algorithms. |
| title | Top-r Influential Community Search in Bipartite Graphs |
| topic | Social and Information Networks |
| url | https://arxiv.org/abs/2412.06216 |