R-ga: an Efficient Method for Predictive Modeling of Medical Data Using a Combined Approach of Random Forests and Genetic Algorithm

Medical data mainly includes data of patients and their associated symptoms. Detecting a disease is becoming costly in terms of money and effort. Medical care will be much better if the predictions can be made with minimal efforts. Predictive modeling will help in detecting a disease early. Medical prediction methods which are computer based will help to improve diagnosis. These methods are the important components of decision support systems. This paper suggests the use of predictive modeling as a classifier. The records in dataset are used to construct classifiers using a combination of random forests and genetic algorithm. The inputs to the predictive model are records from the dataset. Genetic algorithm when used in field of computer science help to form methods that lead to a solution that is acceptable. The results of experimentation show that the random forests when used in combination with genetic algorithm gives better accuracy than random forests algorithm alone.