Abstract
In this paper, we propose a fast method for successive modularity computations in agglomerative clustering for community discovery. Our method is based on a new data structure to maintain two statistical values derived by the definition of modularity between two adjacent (temporary) communities. We prove the correctness and efficiency of the method. Then, we perform it on standard datasets. The experimental results show that our method improves the efficiency significantly.
| Original language | English |
|---|---|
| Pages (from-to) | 153-164 |
| Number of pages | 12 |
| Journal | International Journal of Advancements in Computing Technology |
| Volume | 3 |
| Issue number | 4 |
| DOIs | |
| State | Published - May 2011 |
| Externally published | Yes |
Keywords
- Algorithm
- Community discovery
- Modularity
- Time complexity
Fingerprint
Dive into the research topics of 'Fast computation of modularity in agglomerative clustering methods for community discovery'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver