·414 智能系统学报 第6卷 [65]JEBARA T,WANG J,CHANG S.Graph construction and 2003. b-matching for semi-supervised learing[C]//Proceedings [78]王玲，薄列峰，焦李成.密度敏感的半监督谱聚类[J] of the 26th Annual International Conference on Machine 软件学报，2007,18(10)：2412-2422. Learing.New York,USA:ACM,2009:441448. WANG Ling,BO Liefeng,JIAO Licheng.Density-sensi- [66]DAITCH S I,KELNER J A,SPIELMAN D A.Fitting a tive semi-supervised spectral clustering[J].Journal of graph to vector data[C]//Proceedings of the 26th Annual Software,2007,18(10):2412-2422 Intemational Conference on Machine Learning.New York, [79]KAMVAR S D,KLEIN D,MANNING C D.Spectral USA:ACM,2009:201-208. learing[C]//Proceedings of the Eighteenth Interational [67]XUE F,KUMAR P R.The number of neighbors needed Joint Conference on Artificial Intelligence.Acapulco, for connectivity of wireless networks J].Wireless Net- Mexico,2003:561-566. work8,2004,10(2):169-181. [80]FISCHER I,POLAND I.New methods for spectral cluste- 68]BALISTER P,BOLLOBAS B,SARKAR A,et al.Con- ring,Technical Report No.IDSIA-12-04 [R].Manno, nectivity of random k-nearest-neighbour graphs[J].Ad- Switzerland:Dalle Molle Institute for Artificial Intelli- vances in Applied Probability,2005,37(1):1-24. gence,2004. 69]BRITO M R,CHFIVEZ E L,QUIROZ A J,et al.Con- [81]ROWEIS S,SAUL L.Nonlinear dimensionality reduction nectivity of the mutual k-nearest-neighbor graph in cluste- by locally linear embedding [J].Science,2000,290 ring and outlier detection[J].Statistics Probability Let- (5550):2323-2326. es,1997,35(1):3342. [82]CHENG B,YANG J,YAN S,et al.Learning with L1- [70]POLITO M,PERONA P.Grouping and dimensionality re- graph for image analysis[J].IEEE Transactions on Image duction by locally linear embedding[C]//DIETTERICH T Proce99ing,2010,19(4):858866. G,BECKER S,GHAHRAMANI Z.Advances in Neural 作者简介： Information Processing Systems.Cambridge,USA:MIT 李建元，男，1979年生，讲师，博士 Press,2001:1255-1262. 研究生，CCF及ACM学生会员，主要研 [71]田铮，李小斌，句彦伟.谱聚类的扰动分析[J].中国科 究方向为数据挖掘、机器学习、遥感与 学E辑：信息科学，2007,37(4)：527543 GIS等. TIAN Zheng,LI Xiaobin,JU Yanwei.Perturbation analy- sis of spectral clustering[J].Science in China Series E: Technological Sciences,2007,37(4):527-543. [72]HERNANDEZ V,ROMAN J,TOMAS A,et al.A survey 周脚根，男，1978年生，副研究员， of software for sparse eigenvalue problems EB/OL]. 博士，主要研究方向为空间数据挖掘和 [2010-10-05].http://www.grycap.upv.es/slepc/docu- 空间统计等。 mentation/reports/str6.pdf. [73]FOWLKES C,BELONGIE S,CHUNG F,et al.Spectral grouping using the Nystrom method[J].IEEE Transactions on Pattem Analysis and Machine Intelligence,2004,26 (2):214-225. 关佶红，女，1969年生，教授，博士 [74]DHILLON IS,GUAN Y,KULIS B.Weighted graph cuts 生导师，博士，主要研究方向为分布计 without eigenvectors:a multi-level approach[J].IEEE 算、数据库、数据挖掘、生物信息学等。 Transactions on Pattem Analysis and Machine Intelligence, 2007,29(11):1944-1957. [75]YAN D H,HUANG L,JORDAN M I.Fast approximate spectral clustering[C]//ACM SIGKDD Interational Con- ference on Knowledge Discovery and Data Mining.New 周水庚，男，1966年生，教授，博士 York,USA:ACM Press,2009:907-915. 生导师，博士，主要研究方向为网络数 [76]MASCHHOFF K J,SORENSEN D C.A portable imple- 据管理与搜索、海量数据挖掘与学习、 mentation of ARPACK for distributed memory parallel ar- 生物信息学等。 chitectures[EB/OL].[2010-10-05].http://www.caam. rice.edu/_kristyn/parpack home.html. [77]BACH F R,JORDAN M I.Leaming spectral clustering, Technical Report No.UCB/CSD-03-1249[R].Berkeley, USA:Computer Science Division,University of Califoria