Search Strategies for Binary Feature Selection for a Naive Bayes Classifier
arXiv:1506.04177v1
Originality Synthesis-oriented
AI Analysis
This addresses feature selection for binary data in Naive Bayes Classifiers, but it is incremental as it compares existing methods.
The paper tackled feature selection for Naive Bayes Classifiers with redundant binary data, finding that wrapper methods using classification error probability estimation outperformed filter methods while maintaining reasonable computational cost.
We compare in this paper several feature selection methods for the Naive Bayes Classifier (NBC) when the data under study are described by a large number of redundant binary indicators. Wrapper approaches guided by the NBC estimation of the classification error probability out-perform filter approaches while retaining a reasonable computational cost.