Comparative Study of Different Classification Algorithms on ILPD Dataset to Predict Liver Disorder

Abstract

Data mining techniques can be applied in various fields such as Information Retrieval, Business analytics, Medicine and many more. This paper deals with medical field which mainly focuses on liver disease diagnoses. The aim of this study is to implement different classification algorithms on Indian Liver Patient Dataset (ILPD) using WEKA in order to get proper prediction of liver disorders. Feature selection is carried out on the dataset. Pre-processing is carried out to pre-process and cluster the data. K means clustering algorithm is used for pre-processing the data. The clustered data is further applied to various classification algorithms such as Naive Bayes, Ada Boost, J48, Bagging and Random Forest. A comparison is carried out considering performance measures such as Accuracy, Error rate, Precision, Recall and F measure. On the basis of comparison, the results are concluded. Random Forest algorithm provides best performance among all.