正在加载图片...
第3期 郑文萍,等:基于稠密子图的社区发现算法 ·431. 好的聚类性能。同时,中心社区扩展算法可以有效 CHEN Kehan,HAN Panpan,WU Jian.User clustering 地提高CPM、k-dense算法的聚类性能,该算法也可 based social network recommendation[]].Chinese journal of 用于其他非结点完全覆盖算法。 computers,.2013,36(2):349-359. [10]FREY B J,DUECK D.Clustering by passing messages be- 5结束语 tween data points[J].Science,2007,315(5814):972- 976. 本文提出一种基于稠密子图的图聚类算法 [11]NEWMAN M E J.Community detection and graph partitio- BDSG,解决了基于密度算法中大量未聚类结点问 ning[]Europhysics letters,2013,103(2):28003. 题。通过搜索网络中的相对稠密子图得到中心社 [12]NEWMAN M E J.Spectral methods for community detec- 区:通过定义结点对社区的归属度来度量结点和社 tion and graph partitioning[J].Physical review E,2013, 区连接倾向性,进而给出一种中心社区扩展策略对 88(4):042822. 中心社区外结点进行聚类。通过与CPM、k-dense算 [13]LIN Chuncheng,KANG Jiarong,CHEN J Y.An integer 法在5个真实网络数据集上进行分析比较,结果表 programming approach and visual analysis for detecting hi- 明,BDSG算法在未聚类结点个数、模块性及运行时 erarchical community structures in social networks[J].In- 间方面均表现出较好的性能。同时中心社区扩展策 formation sciences,2015,299:296-311. 略与其他算法相结合,对提高CPM、k-dense等算法 [14]REN Jun,WANG Jianxin,LI Min,et al.Identifying pro- tein complexes based on density and modularity in protein- 的聚类性能具有一定的适用性。 protein interaction network[J].BMC systems biology, 参考文献: 2013,7(S4):S12. [15]LI Xiaoli,FOO C S,NG S K.Discovering protein comple- [1]FORTUNATO S.Community detection in graphs[J].Phys- xes in dense reliable neighborhoods of protein interaction ics reports,2010,486(3/4/5):75-174. networks[C]//Proceedings of the computational systems [2]NEPUSZ T,YU Haiyuan,PACCANARO A.Detecting over- bioinformatics conference.San Diego,USA,2007,6: lapping protein complexes in protein-protein interaction net- 157-168. works[J.Nature methods,2012,9(5):471-472. [16]PALLA G,DEReNYI I,FARKAS I,et al.Uncovering the [3]DEYLAMI H A,ASADPOUR M.Link prediction in social overlapping community structure of complex networks in networks using hierarchical community detectionC//Pro- nature and society J.Nature,2005,435:814-818. ceedings of the 7th conference on information and knowledge [17]SAITO K,YAMADA T,KAZAMA K.Extracting commu- technology.Urmia,Iran,2015:1-5. nities from complex networks by the k-dense method[J]. [4]SCHAEFFER S E.Graph clustering[J].Computer science IEICE transactions on fundamentals of electronics,commu- review,2007,1(1):27-64. nications and computer sciences,2008,E91-A (11 ) [5]杨博,刘大有,U Jiming,等.复杂网络聚类方法[J]. 3304-3311. 软件学报,2009,20(1):54-66. [18]SUN Penggang,GAO Lin.Fast algorithms for detecting o- YANG Bo,LIU Dayou,LIU Jiming,et al.Complex network verlapping functional modules in protein-protein interaction clustering algorithms [J].Journal of software,2009,20 networks[C]//Proceedings of the IEEE computational in- (1):54-66. telligence in bioinformatics and computational biology. [6]冀俊忠,刘志军,刘红欣,等.蛋白质相互作用网络功 Nashville,TN,USA,2009:247-254. 能模块检测的研究综述[J].自动化学报,2014,40(4): [19]LIU Guimei,WONG L,CHUA H N.Complex discovery 577.593. from weighted PPI networks[J].Bioinformatics,2009,25 JI Junzhong,LIU Zhijun,LIU Hongxin,et al.An overview (15):1891-1897. of research on functional module detection for protein-protein [20]BADER G D,HOGUE C W V.An automated method for interaction networks J.Acta automatica sinica,2014,40 finding molecular complexes in large protein interaction (4):577-593 networks[.BMC bioinformatics,2003,4(1):2. [7]PALLA G,BARABASI A L,VICSEK T.Quantifying social [21]FREEMAN L C.A set of measures of centrality based on group evolution[J].Nature,2007,446(7136):664-667. betweenness[J].Sociometry,1977,40(1):35-41. [8]SIDIROPOULOS A,PALLIS G,KATSAROS D,et al.Pre- [22 CUI Yaizu,WANG Xingyuan,EUSTACE J.Detecting fetching in content distribution networks via web communi- community structure via the maximal sub-graphs and be- ties identification and outsourcing[J].World wide web, longing degrees in complex networks[J].Physica A:sta- 2008,11(1):39-70. tistical mechanics and its applications,2014,416:198- [9]陈克寒,韩盼盼,吴健.基于用户聚类的异构社交网络 207. 推荐算法[J].计算机学报,2013,36(2):349-359. [23]ZACHARY WW.An information flow model for conflict好的聚类性能。 同时,中心社区扩展算法可以有效 地提高 CPM、k⁃dense 算法的聚类性能,该算法也可 用于其他非结点完全覆盖算法。 5 结束语 本文提出一种基于稠密子图的图聚类算法 BDSG,解决了基于密度算法中大量未聚类结点问 题。 通过搜索网络中的相对稠密子图得到中心社 区;通过定义结点对社区的归属度来度量结点和社 区连接倾向性,进而给出一种中心社区扩展策略对 中心社区外结点进行聚类。 通过与 CPM、k⁃dense 算 法在 5 个真实网络数据集上进行分析比较,结果表 明,BDSG 算法在未聚类结点个数、模块性及运行时 间方面均表现出较好的性能。 同时中心社区扩展策 略与其他算法相结合,对提高 CPM、k⁃dense 等算法 的聚类性能具有一定的适用性。 参考文献: [1]FORTUNATO S. Community detection in graphs[ J]. Phys⁃ ics reports, 2010, 486(3 / 4 / 5): 75⁃174. [2]NEPUSZ T, YU Haiyuan, PACCANARO A. Detecting over⁃ lapping protein complexes in protein⁃protein interaction net⁃ works[J]. Nature methods, 2012, 9(5): 471⁃472. [3]DEYLAMI H A, ASADPOUR M. Link prediction in social networks using hierarchical community detection[C] / / Pro⁃ ceedings of the 7th conference on information and knowledge technology. Urmia, Iran, 2015: 1⁃5. [4]SCHAEFFER S E. Graph clustering[ J]. Computer science review, 2007, 1(1): 27⁃64. [5]杨博, 刘大有, LIU Jiming, 等. 复杂网络聚类方法[ J]. 软件学报, 2009, 20(1): 54⁃66. YANG Bo, LIU Dayou, LIU Jiming, et al. Complex network clustering algorithms [ J ]. Journal of software, 2009, 20 (1): 54⁃66. [6]冀俊忠, 刘志军, 刘红欣, 等. 蛋白质相互作用网络功 能模块检测的研究综述[J]. 自动化学报, 2014, 40(4): 577⁃593. JI Junzhong, LIU Zhijun, LIU Hongxin, et al. An overview of research on functional module detection for protein⁃protein interaction networks[J]. Acta automatica sinica, 2014, 40 (4): 577⁃593. [7]PALLA G, BARABáSI A L, VICSEK T. Quantifying social group evolution[J]. Nature, 2007, 446(7136): 664⁃667. [8]SIDIROPOULOS A, PALLIS G, KATSAROS D, et al. Pre⁃ fetching in content distribution networks via web communi⁃ ties identification and outsourcing [ J]. World wide web, 2008, 11(1): 39⁃70. [9]陈克寒, 韩盼盼, 吴健. 基于用户聚类的异构社交网络 推荐算法[J]. 计算机学报, 2013, 36(2): 349⁃359. CHEN Kehan, HAN Panpan, WU Jian. User clustering based social network recommendation[J]. Chinese journal of computers, 2013, 36(2): 349⁃359. [10]FREY B J, DUECK D. Clustering by passing messages be⁃ tween data points[ J]. Science, 2007, 315( 5814): 972⁃ 976. [11]NEWMAN M E J. Community detection and graph partitio⁃ ning[J]. Europhysics letters, 2013, 103(2): 28003. [12]NEWMAN M E J. Spectral methods for community detec⁃ tion and graph partitioning[ J]. Physical review E, 2013, 88(4): 042822. [13]LIN Chuncheng, KANG Jiarong, CHEN J Y. An integer programming approach and visual analysis for detecting hi⁃ erarchical community structures in social networks[ J]. In⁃ formation sciences, 2015, 299: 296⁃311. [14]REN Jun, WANG Jianxin, LI Min, et al. Identifying pro⁃ tein complexes based on density and modularity in protein⁃ protein interaction network [ J ]. BMC systems biology, 2013, 7(S4): S12. [15]LI Xiaoli, FOO C S, NG S K. Discovering protein comple⁃ xes in dense reliable neighborhoods of protein interaction networks[ C] / / Proceedings of the computational systems bioinformatics conference. San Diego, USA, 2007, 6: 157⁃168. [16]PALLA G, DERéNYI I, FARKAS I, et al. Uncovering the overlapping community structure of complex networks in nature and society[J]. Nature, 2005, 435: 814⁃818. [17]SAITO K, YAMADA T, KAZAMA K. Extracting commu⁃ nities from complex networks by the k⁃dense method[ J]. IEICE transactions on fundamentals of electronics, commu⁃ nications and computer sciences, 2008, E91⁃A ( 11 ): 3304⁃3311. [18]SUN Penggang, GAO Lin. Fast algorithms for detecting o⁃ verlapping functional modules in protein⁃protein interaction networks[C] / / Proceedings of the IEEE computational in⁃ telligence in bioinformatics and computational biology. Nashville, TN, USA, 2009: 247⁃254. [19] LIU Guimei, WONG L, CHUA H N. Complex discovery from weighted PPI networks[J]. Bioinformatics, 2009, 25 (15): 1891⁃1897. [20]BADER G D, HOGUE C W V. An automated method for finding molecular complexes in large protein interaction networks[J]. BMC bioinformatics, 2003, 4(1): 2. [21]FREEMAN L C. A set of measures of centrality based on betweenness[J]. Sociometry, 1977, 40(1): 35⁃41. [22] CUI Yaizu, WANG Xingyuan, EUSTACE J. Detecting community structure via the maximal sub⁃graphs and be⁃ longing degrees in complex networks[J]. Physica A: sta⁃ tistical mechanics and its applications, 2014, 416: 198⁃ 207. [23] ZACHARY W W. An information flow model for conflict 第 3 期 郑文萍,等:基于稠密子图的社区发现算法 ·431·
<<向上翻页向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有