Thyroid Classification using Ensemble Model with Feature Selection

Now days, diagnosis of health conditions is a very critical and challenging task in field of medical science. Medical history data comprises of a number of tests essential to diagnose a particular disease and the diagnoses are based on the physician experience. The thyroid gland faced by physician which is one of the important organs in the body and also increases cellular activity. Data mining technique can greatly deal the diseases of patients. Classification is one of the most important decision making techniques in many real world problem. In this paper, the main objective is to classify the data as thyroid or non thyroid and improve the classification accuracy. We have used various classifications techniques and its ensemble model for classification of thyroid data. Feature selection technique is also important role to improve the classification accuracy and increases performance. An ensemble of C4.5 and Random forest gives better accuracy 99.47% with 5 features.