DEVELOPMENT OF R-CODI METHOD FOR OVERLAPPING COMMUNITIES DETECTION
Over the past 10 years, data exploration on social media networks has become a very popular topic to study. One interesting knowledge that can be found from social media networks is community. Many methods have been developed to detect communities. One of them is called R-CoDi method (Random-based C...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/36803 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Over the past 10 years, data exploration on social media networks has become a very popular topic to study. One interesting knowledge that can be found from social media networks is community. Many methods have been developed to detect communities. One of them is called R-CoDi method (Random-based Community Diffusion). The R-CoDi method is a community detection method that does not require structural parameters from social media networks. This method works by using data generated from the process of spreading a contagion over the network.
Based on the results of the analysis, it can be concluded that the R-CoDi method and software only detect one community for each node, unstable because the number of communities and community members varies with each execution, still detect nodes that are not infected by contagion, and still read input statically.
In this paper, the author developed a method and software called R-CoDi ++. Based on the test results, the R-CoDi++ method and software have optimized four deficiencies of the R-CoDi method and software that had been mentioned above. |
|---|