ANALYSIS OF IMPACT OF FUZIFICATION OF RANGE ATTRIBUTES IN PERFORMANCE OF VARIOUS CLASSIFICATION ALGORITHMS IN DETECTING PRIMARY TUMOR

Thendral Puyalnithi
2016 International Journal of Pharmacy & Technology   unpublished
The performance of the classifiers can be enhanced in various ways, one way of doing is changing the attribute nature, by fuzifying the attributes which are of range type, considerable increase in the precision of basic classification algorithms are found. Comparison of performance of ID3, SVM, KNN, Random Forest and Naive Bayes algorithms tested with data set with an without fuzified values and the results are analysed. The results supports the idea that the fuzification of data will increase the efficiency of classifiers.
fatcat:4qadxngtjbdrddb7u4l2l2ghwy