Disease diagnosis through data mining techniques

Data mining is the process of extracting information or discovering usable data from complex database. Medical field contains huge heterogeneous data which is being converted into useful information by applying data mining techniques and then this useful data is utilized by the physicians to diagnose various diseases with satisfactory accuracy. In this work, authors have been focused on the diagnosis of thyroid diseases by using neural net, vote ensemble and stacking ensemble methods, where the classifiers are compared in terms of accuracy, precision, recall, and error rate. Experimental result shows that Stacking ensemble method has highest accuracy than any other methods and also has better prediction accuracy when compared with the related existing literature.