利用改进蚁群算法的重叠社团检测分析方法

针对重叠社团检测准确率提升问题，提出了一种基于改进蚁群算法的新型重叠社团检测算法。该算法包含位置初始化、运动和后处理三个阶段，分别通过初始位置识别与标签列表存储、基于节点间相似度的启发式信息重定义、合作保持标签列表等方式，使算法在合成数据集与现实世界数据集中的重叠社团与节点检测方面具有更好的性能。实验结果表明，在合成网络与现实世界网络平台上使用不同检测算法，所提出的方法对重叠社团与重叠节点的检测准确率较传统检测方法来说更高，因而对重叠社区检测问题求解与理解网络功能结构具有重要的参考与借鉴意义。

Novel algorithm of overlapping community detection and analysis with improved ant colony algorithm

This paper proposed a new detection algorithm for overlapping communities based on ant colony algorithm to improve the detection accuracy of overlapping communities. The algorithm consisted of three stages: position initialization, motion and post-processing. The algorithm achieved better performance of overlapping communities and overlapping nodes detection in synthetic datasets and real-world datasets through initial position identification and tag list storage, heuristic information redefinition based on similarity between nodes, and cooperative tag list preservation. The detection performance of different detection algorithms on synthetic and real world network platforms shows that the proposed method based on improved ant colony algorithm has good accuracy and analysis performance. The method can be used for reference in solving the current overlapping community detection problem and understanding the functional structure of the network.