Overlapping Community Detection Based on Strong Tie Detection and Non-Overlapping Link Clustering

Many clustering algorithms are in favour of node-based methods, but a link between nodes has one single feature, so link-based is sometimes easier than methods. Being dependent on the characteristics links, detection algorithm for non-overlapping community put forward this paper. The method proposed also distinguishes differences with high degree accuracy and detects communities minimal number overlapping nodes. On basis three different datasets, experiments were conducted to compare algorithms, results show that our generates least achieves good partition.