Community detection is the problem of identifying densely connected clusters of nodes within a network. The Louvain algorithm is a widely used method for this task, but it can produce communities that are internally disconnected. To address this, the Leiden algorithm was introduced. In this technical report, we propose another approach to mitigate this issue. On a system with two 16-core Intel Xeon Gold 6226R processors, our new parallel algorithm GSP-Louvain, based on the Louvain algorithm, addresses this issue, and outperforms the original Leiden, igraph Leiden, and NetworKit Leiden by 341x, 83x, and 6.1x respectively - achieving a processing rate of 328M edges/s on a 3.8B edge graph. Furthermore, GSP-Louvain improves performance at a rate of 1.5x for every doubling of threads.
翻译:暂无翻译