GA_DTNB: A Hybrid Classifier for Medical Data Diagnosis

摘要

Recent trends in medical data prediction have become one of the most challenging tasks for the researchers due to its domain specificity, voluminous, and class imbalanced nature. This paper proposed a genetic algorithms (GA)-based hybrid approach by combining decision table (DT) and Na?ve Bayes (NB) learners. The proposed approach is divided into two phases. In the first phase, feature selection is done by applying GA search. In the second phase, the newly obtained feature subsets are used as input to combined DTNB to enhance the classification performances of medical data sets. In total, 14 real-world medical domain data sets are selected from University of California, Irvine (UCI) machine learning repository, for conducting the experiment. The experimental results demonstrate that GA-based DTNB is an effective hybrid model in undertaking medical data prediction.