Compaative Analysis of Random Forest And Logistic Regression For Diagnosis Of Diabetes Mellitus

Abstract

In our daily life there is lots of data in different field. Whenever there is data we can have lots of information, patterns, meaning etc. and the process of Extracting or “mining” knowledge from large amount of data is called Data mining and is also known as “Knowledge discovery from data (KDD)”. Data mining applications has got rich focus due to its significance of classification algorithms. Diabetes Mellitus (DM) is a result of bad metabolism. DM, if not controlled, causes several complications and even affects other parts of the body. This study aims to survey on the two different classifiers with data set of patients regarding Diabetes Mellitus and to implement as well as assist by comparing Random Forest and Logistic Regression classification techniques to standardize the diagnosis and treatment of Diabetes Mellitus. From the context analysis it was seen that Logistic Regression was able to classify 81.17% of the data correctly which was better than Random Forest in comparison to results of evaluation metrics (Accuracy, Precision, Recall and F-Measure). In a nut shell, the experiment result showed that Logistic Regression had got 2% better accuracy than Random Forest for the diagnosis of diabetes mellitus.