·480. 智能系统学报 第7卷 nosis for diesel value trains based on non-negative matrix fac- WANG Chao,WANG Shitong.Maximum margin and lo- torization and neural network ensemble[J].Mechanical Sys- cality preserving feature selection method[J].Artificial In- tems and Signal Processing,2009,23(5):1683-1695. telligence and Pattern Recognition,2009,22(6):898- [2]CAMACHO J,PIC J,FERRER A.Data understanding with 902. PCA:structural and variance information plots[J].Chemomet- [14]ODIOWEI PP,YI Cao.Nonlinear dynamic process moni- rics and Intelligent Laboratory Systems,2010,100(1):48-56. toring using canonical variate analysis and kemel density [3]LIPOVETSKY S.PCA and SVD with nonnegative loadings estimations J].IEEE Trans on Industrial Informatics, [J].Pattern Recognition,2009,42(1):68-76. 2010,6(1):3645. [4]RADULOVIC J,RANKOVIC V.Feedforward neural net- [15]吴葛铭，霍剑青，王晓蒲.一种基于加权Parzen窗的聚 work and adaptive network-based fuzzy inference system in 类算法[J].中国科学技术大学学报，2002,32(5)： study of power lines[J].Expert Systems with Applications, 546551. 2010,37(1):165-170. WU Geming,HUO Jianging,WANG Xiaopu.Weighted [5]PETER N B,JOAO P H,DAVID J K.Eigenfaces vs Parzen window based clustering[J].Journal of Chinease Uni- Fisherfaces:recognition using class specific linear projec- versity of Sciences and Technologies,2002,32(5):546-551. tion[J].IEEE Trans on Pattern Analysis and Machine In- [16]HE X F,YAN S C,HU Y X,et al.Face recognition u- telligence,1997,19(7):711-720. sing Laplacianfaces[J].IEEE Trans on Pattern Analysis [6]HU Y H,HE S H.Integrated evaluation method[M].Bei- and Machine Intelligence,2005,27(3):328-340. jing:Scientific Press,2000:27-31. [17]CAMACHO J,PIC J,FERRER A.Data understanding [7]ROWEIS S T,SAUL L K.Nonlinear dimensionality reduc- with PCA:structural and variance information plots[J]. tion by locally linear embedding[J].Science,2000(290): Chemometrics and Intelligent Laboratory Systems,2010, 2323-2326. 100(1)：48-56. [8]BELKIN M,IYOGI P.Laplacian eigenmaps and spectral tech- [18]边肇祺，张学工.模式识别[M].2版.北京清华大学 niques for embedding and clustering[C]/Systems (NIPS). 出版社，2000：1-30. Cambridge:MIT NIYOGI P Press,2001:585-591. [19 ]GUO Q C,CHANG X J,CHU H X.Mean-shift of varia- [9]BORGA M,KNUTSSON H.Canonical correlation analysis ble window based on the Epanechnikov kemel[C]//Pro- in early vision[C]//Proceedings of the 9th European Sym- ceedings of IEEE Int Conf on Mechatronics and Automa- posium on Artificial Neural Networks.Bruges,Belgium, tion.Piscataway,USA,2007:2314-2319. 2001:309-314. 作者简介： [10]HE X F,NIYOGI P.Locality preserving projections[C]// 刘忠宝，男，1981年生，博士，主要 Advances in Neural Information Processing Systems 研究方向为模式识别、机器学习，发表 (NIPS).Vancouver,Canada,2003:153-160. 学术论文数十篇. [11]皋军，王士同，邓赵红.广义的势支撑特征选择GPS FM[J].计算机研究与发展，2009,46(1)：41-51 GAO Jun,WANG Shitong,DENG Zhaohong.Generalized potential support vector machine for feature selection[]. Computer Research and Development,2009,46(1):41-51. [12]王晓明，王士同.广义的监督局部保留投影算法[J] 王士同，男，1964年生，教授，博士 电子与信息学报，2009,31(8)：1840-1845. 生导师，主要研究方向为人工智能与机 WANG Xiaoming,WANG Shitong.Generalized supervised 器学习. locality preserving projection algorithm[].Electronics and Information,2009,31(8):1840-1845. [13]王超，王士同.有局部保持的最大间距准则特征提取 方法[J].模式识别与人工智能，2009,22(6)：898- 902