This project implements the paper [1] that proposes a random-walk based algorithm Walktrap to efficiently find clusters in graphs. Also, I add extensive comparison tests with other algorithms and results from experiments on randomly-generated as well as real-world networks. One additional community detection algorithm and two real-world networks beyond the scope of the original paper are considered.
- Report [PDF]
[1] Pascal Pons and Matthieu Latapy. Computing communities in large networks using random walks. J. Graph Algorithms Appl., 10(2):191โ218, 2006.