CAJ | 학술논문

复杂网络的社团发现问题是网络数据挖掘中的重要问题之一。利用基于模糊 C 均值的细菌群体趋药性算法最大化网络的模块度，算法中模糊 C 均值的初始值由群体细菌取药性算法获得。模糊 C 均值算法在此基础上发现复杂网络的社团结构。其创新点在于最佳模块度的寻找。实验结果表明：该算法具有对现实世界网络社团划分的可行性和有效性。
복잡망락적사단발현문제시망락수거알굴중적중요문제지일。이용기우모호 C 균치적세균군체추약성산법최대화망락적모괴도，산법중모호 C 균치적초시치유군체세균취약성산법획득。모호 C 균치산법재차기출상발현복잡망락적사단결구。기창신점재우최가모괴도적심조。실험결과표명：해산법구유대현실세계망락사단화분적가행성화유효성。
Identification of communities in a complex network is one of the important problems in data min‐ing of network data .The bacterial colony chemotaxis (BCC) strategy with fuzzy C‐means (FCM ) algorithm was used to maximize the modularity of a network .In the new algorithm ,the initial cluster center of FCM algorithm was obtained by BCC algorithm .Then ,the FCM algorithm was used for detecting communities in a complex network .The proposed algorithm outperformed most the existing methods in the literature as regards the opti‐mal modularity found .Experimental results for real‐word networks confirmed the feasibility and effectiveness of the proposed algorithm .