Hybrid Approach for Diagnosing Thyroid, Hepatitis, and Breast Cancer Based on Correlation Based Feature Selection and Naïve Bayes

Feature selection techniques have become an obvious need for researchers in computer science and many other fields of science. Whether the target research is in medicine, agriculture, business, or industry; the necessity for analysing large amount of data is needed. In Addition to that, finding the most excellent feature selection technique that best satisfies a certain learning algorithm could bring the benefit for researchers. Therefore, we proposed a new method for diagnosing some diseases based on a combination of learning algorithm tools and feature selection techniques. The idea is to obtain a hybrid approach that combines the best performing learning algorithms and the best performing feature selection techniques in regards to three well-known datasets. Experimental result shows that co-ordination between correlation based feature selection method along with Naive Bayse learning algorithm can produce promising results.