Kernel k-nearest neighbor algorithm as a flexible SAR modeling tool
作者单位：Research center of modernization of traditional Chinese medicines Central South UniversityChangsha 410083 P.R. China School of Mathematical Sciences and Computing Technology Central South UniversityChangsha 410083 P.R. China
会议名称：《中国化学会第28届学术年会》
会议日期：2012年
学科分类：081704[工学-应用化学] 07[理学] 070304[理学-物理化学(含∶化学物理)] 08[工学] 0817[工学-化学工程与技术] 0703[理学-化学]
关 键 词：k-nearest neighbor (k-NN) kernel methods string kernel structure-activity relationship (SAR).
摘 要：A kernel version of k-nearest neighbor algorithm （k-NN） has been developed to model the complex relationship between molecular descriptors and bioactivities of *** k-NN is to perform the original k-NN algorithm by mapping the training samples in the input space into a high-dimensional feature *** can be easily constructed by calculating the distance between samples in the feature space,directly deriving from the simple calculation of the kernel *** developed kernel k-NN is very flexible to deal with complex nonlinear relationship,more importantly;it can also conveniently cope with some non-vectorial data only by the definition of different *** results obtained from several real SAR datasets indicated that the performance of kernel k-NN is comparable to support vector machine *** can be regarded as an alternative modeling technique for several chemical problems including the study of structure-activity relationship （SAR）.The source codes implementing kernel k-NN in R language are freely available at http://***/p/kernelmethods/.