In this work we propose a method for computing a minimum size training set consistent subset for the Nearest Neighbor rule (also said CNN problem) via SAT encodings. We introduce the SAT-CNN algorithm, which exploits a suitable encoding of the CNN problem in a sequence of SAT problems in order to exactly solve it, provided that enough computational resources are available. Comparison of SAT-CNN with well-known greedy methods shows that SAT-CNN is able to return a better solution. The proposed approach can be extended to several hard subset selection classification problems. © 2008 International Federation for Information Processing.
CITATION STYLE
Angiulli, F., & Basta, S. (2008). Optimal subset selection for classification through SAT encodings. In IFIP International Federation for Information Processing (Vol. 276, pp. 309–318). https://doi.org/10.1007/978-0-387-09695-7_30
Mendeley helps you to discover research relevant for your work.