参考文献/References:
[ 1] Sch lkop f B, Smo laA. Learn ingW ith Kerne ls[M ] . C amb ridge, MA: M IT Press, 2002.
[ 2] Shaw e-Tay lo r J, Cr istian in iN. K ernelM ethods for Pattern Ana ly sis[M ]. Cambr idg e: Cambr idg eUn iversity Press, 2004.
[ 3] Vapn ik V. Sta tistica l Lea rning Theo ry[M ]. New York: Add ison-W iley, 1998.
[ 4] Cr istian ini N, T ay lor J S. An Introduction to Support Vecto rM achines and O ther Kerne-l based Learn ing M ethods[M ]. C ambridge: Cambr idge Un iversity Press, 2000.
[ 5] M ika S, R tsch G, W eston J, et a.l F isher d iscr im inant ana lysis w ith kerne ls[ C ] / / Neura lNetw orks fo r Signa lProcessing IX.Piscataw ay, NJ: IEEE, 1999.
[ 6] Sch lkopf B, Sm olaA J, M llerK R. Nonlinear com ponen t analysis as a kerne l e ig envalue problem [ J]. N eural Com puta tion,1998( 10): 1 299-1 319.
[ 7] Sch lkop f B, W illiam son R C, Sm o la A J. Support vectorm e thod for novelty detection[ C] / / Advances in N eural Inform ation Process ing System s. Cambr idge: M IT Press, 2000.
[ 8] Ta rassenko L, H ay ton P, BradyM. Nove lty de tection for the identification o f masses in m amm ogram s[ C] / / Proc 4th Int IEE Con fA rtif Neura lNe tw. Cam bridge: Ox ford University Press, 1995.
[ 9] La zarev icA, Erto z L, Kum arV, e t a.l A comparative study o f anom aly detection schem es in ne tw ork intrusion detection[ C ] / /SDM 2003. San Francisco: S IAM, 2003.
[ 10] Roth V. Outlier detection w ith one-c lass kerne l fisher discr im inants[ C ] / / Adv ances in Neura l Inform ation Process ing Systems. Cam bridge: M IT Press, 2005.
[ 11] Chen Y, Zhou X, H uang T. One- class SVM fo r learn ing in image re trieval[ J]. Im age Processing, 2001( 1): 34-37.
[ 12] M anev itz L, Youse fM. One- class SVM s for docum ent c lassification[ J] . Journa l ofM ach ine Learn ing Research, 2001( 2):139-154.
[ 13] M arkou M, S ingh S. Nove lty detection: a rev iew-part 1: statistica l approaches[ J] . S ignal Processing, 2003, 83( 12): 2481-2 497.
[ 14] MoyaM, KochM, H ostetler L. One-c lass class ifier netwo rks fo r ta rget recogn ition app lications[ C] / /Proceed ingsW orld Congress
on Neura lNe tw orks. Portland, OR: Internationa lNeural Netwo rk Soc ie ty, 1993: 797-801.
[ 15] Juszczak P. Learn ing to recogn ise: a study on one-c lass c lassiifcation and active learn ing [ D ]. De lft: De lft University o f Techno logy, 2006.
[ 16] Tax D. One-c lass classification: concept- lea rning in the absence o f counter- examp les[ D]. De lft: DelftUn iversity of Techno-logy, 2001.
[ 17] Vapnik V N. The Nature o f Statistica lLearn ing Theory[M ] . New Yo rk: Spr ing er-Ve rlag, 1995.
[ 18] Sch lkopf B, P la tt J C, Shaw e-Taylor J. Estim ating the suppo rt of a h igh-d im ensiona l d istribution[ J]. Neura l Compu tation,2001, 13( 7): 1 443-1 471.
[ 19] Sch lkopf B, Sm ola A J, W illiam son R C, et a .l New support vecto r algor ithm s[ J] . Neural Com putation, 2000, 12 ( 5):1 207-1 245.
[ 20] P latt J. Fast tra in ing of suppo rt vector m ach ines using sequentia lm in im al optim ization[ C ] / / Advances in K ernelM ethods-Suppo rtV ector Learn ing. Cambr idg e: M IT Press, 1999.
[ 21] Tax D, Duin R P. Suppo rt vector doma in descr iption[ J]. Pa ttern Recognition Letters, 1999, 20( 11-13): 191-1 199.
[ 22] Tax D, Duin R P. Suppo rt vector da ta desc ription[ J]. M ach ine Learn ing, 2004, 54( 1) : 45-66.
[ 23] Scho lkopf B, P latt J, Sm o laA. Ke rnelm ethod fo r pe rcentile feature extraction, M SR- TR - 2000- 22[ R]. M icrosoft Technical Report, 2000.
[ 24] Campbe ll C, Bennett K P. A linea r programm ing approach to nove lty detection[ C ] / / Advances in N eural Info rm ation Processing
Sy stem s. Cam bridge: M IT Press, 2001.
[ 25] 冯爱民, 陈斌. 基于局部密度的单类分类器LP改进算法[ J]. 南京航空航天大学学报, 2006, 38( 6) : 727-731.
Feng A im in, Chen B in. Im prov ing LP a lgo rithm s o f one-class class ifier based on the loca l density fac to r[ J] . Journa l o fNanjing
University of Aeronautics and Astronautics, 2006, 38( 6): 727-731. ( in Chinese)
[ 26] A lbertoM, JM. One-c lass support vector m ach ines and density estim ation: the prec ise re lation[ C ] / / Prog ress in Pattern Recogn ition, LNCS. Ber lin: Springer, 2004.
[ 27] Lanckriet G R G, Ghaou iL E, Bhattacharyya C, et a.l A robustm in im ax approach to classification[ J]. Journa l ofM achine Learn ing Research, 2002( 3) : 555582.
[ 28] Tsang IW, Jam es T K, Li S. Learn ing the kerne l inM aha lanob is one- class support vectorm ach ines[ C] / / Pro ceedings o f the
In ternational Joint Conference on Neura l Netwo rks. Canada: Vancouver, 2006.
[ 29] Sch lkop B. K ernel me thods for imp lic it surface m ode ling[ C ] / / Advances in Neural Inform ation Process ing System s. Vancouver,British Co lum bia, Canada: N IPS, 2004.
[ 30] Tao Q, W u G W, W ang J. A new m ax im um m arg in a lgor ithm fo r one-c lass problem s and its boosting imp lementation [ J].
Pattern Recognition, 2005, 38( 77) : 1 071-1 077.
[ 31] DoliaA, H arris C, Shaw e-Tay lor J K, e t a .l Ke rnel ellipso ida l trimm ing [ J]. Com putationa l Statistics and Data Ana lys is,2007, 52( 1): 309-324.
[ 32] DoliaA N, B ie T D, H arris C J, et a.l The m in imum vo lum e cove ring e llipso id estim ation in kerne-l defined feature spaces
[ C ] / / Proc of the 17th European Conference onM achine Learning. Ber lin: Springer-Verlag, 2006.
[ 33] Langford J, Shaw e-Tay lor J. PAC Bayes and m arg ins[ C ] / / Advances in Neural Inform ation Processing System s. Vancouver andWh istler, B ritish Co lumb ia: M IT Press, 2003.
[ 34] W ang D, Yeung D S, Tsang E C C. Structured one-c lass c lassifica tion[ J] . IEEE Trans on System s, M an, and Cybernetics-Part B: Cybernetics, 2006, 36( 6): 1 283-1 294.
[ 35] W e iX K, H uangG B, LiY H. M ahalanobis e llipso idal learn ingm ach ine for one c lass classification[ C] / / Pro ceedings o f the 6 th Internationa l Con ference onM ach ine Learn ing and Cyberne tics. H ong Kong: IEEE Press, 2007.
[ 36] Tax D, Juszczak P. Kerne lwh itening fo r one-c lass c lassifica tion[ J] . Internationa l Journa l of Pa ttern Recognition and Artificial Inte lligence, 2003, 17( 3) : 333-347.
[ 37] H o ffm ann H. Kerne l PCA for nove lty detection[ J]. Pa ttern Recognition, 2007, 40: 863-874.