Call for NIPS 2008 Kernel Learning Workshop Submissions
Document Actions
I. Guyon, N. Matic, and V. Vapnik (1996)
Discovering informative patterns and data cleaning.
In: Advances in Knowledge Discovery and Data Mining, ed. by U.M. Fayyad and G. Piatetsky-Shapiro and P. Smyth and R. Uthurusamy, pp. 181–203.
SVM's are used as a method for discovering informative patterns such that large databases can be reduced to only few representative data entries. This framework also encompasses methods for cleaning databases containing corrupted data. Experiments on handwriting data are presented.