基于連邊距離矩陣的重疊社區(qū)發(fā)現(xiàn)
發(fā)布時(shí)間:2019-07-16 09:51
【摘要】:現(xiàn)有重疊社團(tuán)發(fā)現(xiàn)算法大多直接從相鄰連邊的相似性出發(fā),不能有效利用網(wǎng)絡(luò)的多層連邊信息,基于此提出了一種基于連邊距離矩陣的重疊社區(qū)發(fā)現(xiàn)算法LDM。首先結(jié)合連邊—節(jié)點(diǎn)—連邊隨機(jī)游走模型,以實(shí)現(xiàn)多級連邊信息的有效利用;借助模糊聚類方法,處理連邊距離矩陣以獲取連邊社區(qū);最后根據(jù)擴(kuò)展模塊度調(diào)整和優(yōu)化重疊社區(qū)結(jié)構(gòu)。在人工網(wǎng)絡(luò)和真實(shí)網(wǎng)絡(luò)上的實(shí)驗(yàn)結(jié)果表明,所提算法能夠有效提高重疊社區(qū)發(fā)現(xiàn)算法的準(zhǔn)確度。
[Abstract]:Most of the existing overlapping community discovery algorithms start directly from the similarity of adjacent connected edges, and can not effectively make use of the multi-layer connected edge information of the network. Based on this, an overlapping community discovery algorithm LDM. based on connected edge distance matrix is proposed. Firstly, the connected edge-node-edge random walk model is combined to realize the effective utilization of multi-level connected edge information; with the help of fuzzy clustering method, the connected edge distance matrix is processed to obtain the connected edge community; finally, the overlapping community structure is adjusted and optimized according to the extended modularity. The experimental results on artificial network and real network show that the proposed algorithm can effectively improve the accuracy of overlapping community discovery algorithm.
【作者單位】: 國家數(shù)字交換系統(tǒng)工程技術(shù)研究中心;
【基金】:國家“973”計(jì)劃資助項(xiàng)目(2012CB315901,2012CB315905) 國家自然科學(xué)基金創(chuàng)新群體項(xiàng)目(61521003)
【分類號】:O157.5
本文編號:2515003
[Abstract]:Most of the existing overlapping community discovery algorithms start directly from the similarity of adjacent connected edges, and can not effectively make use of the multi-layer connected edge information of the network. Based on this, an overlapping community discovery algorithm LDM. based on connected edge distance matrix is proposed. Firstly, the connected edge-node-edge random walk model is combined to realize the effective utilization of multi-level connected edge information; with the help of fuzzy clustering method, the connected edge distance matrix is processed to obtain the connected edge community; finally, the overlapping community structure is adjusted and optimized according to the extended modularity. The experimental results on artificial network and real network show that the proposed algorithm can effectively improve the accuracy of overlapping community discovery algorithm.
【作者單位】: 國家數(shù)字交換系統(tǒng)工程技術(shù)研究中心;
【基金】:國家“973”計(jì)劃資助項(xiàng)目(2012CB315901,2012CB315905) 國家自然科學(xué)基金創(chuàng)新群體項(xiàng)目(61521003)
【分類號】:O157.5
【相似文獻(xiàn)】
相關(guān)期刊論文 前7條
1 張福振;;關(guān)于距離矩陣的最佳歐氏逼近[J];北京師范大學(xué)學(xué)報(bào)(自然科學(xué)版);1987年04期
2 楊秀文;陳振杰;李愛玲;田艷芳;;利用矩陣翻轉(zhuǎn)法求最佳H圈[J];后勤工程學(xué)院學(xué)報(bào);2008年01期
3 王忠義;;矩陣的一種特殊運(yùn)算及其應(yīng)用[J];數(shù)學(xué)的實(shí)踐與認(rèn)識;2012年03期
4 丁樹良;;成分?jǐn)?shù)據(jù)統(tǒng)計(jì)分析中變異矩陣若干性質(zhì)及應(yīng)用[J];江西師范大學(xué)學(xué)報(bào)(自然科學(xué)版);1992年04期
5 徐柱;;數(shù)學(xué)分類學(xué)基本知識(四)[J];中國草原;1985年04期
6 陳伯成;利用距離矩陣求絕對中心及絕對重心的討論(p=1)[J];系統(tǒng)工程理論與實(shí)踐;1997年01期
7 謝治州;;突發(fā)事件中對嫌疑犯有效封堵的模型與算法[J];數(shù)學(xué)的實(shí)踐與認(rèn)識;2013年01期
,本文編號:2515003
本文鏈接:http://sikaile.net/kejilunwen/yysx/2515003.html
最近更新
教材專著