CHRONIC KIDNEY DISEASE ANALYSIS USING DATA MINING CLASSIFICATION TECHNIQUES AS A MACHINE LEARNING PERSPECTIVE

Chronic Kidney Disease (CKD) means the kidneys are damaged and are not filtering the blood the way they should. The primary role of kidneys is to filter extra water and waste from the blood to produce urine. If a person suffers from CKD, it means the wastes are collected inside the body which is chronic and can gradually cause damage to the kidneys in the long run. There are many causes for CKD like diabetes, high blood pressure, heart disease. If a particular kidney is not working, then one may notice one or more symptoms like abdominal pain, back pain, diarrhoea, fever, nosebleeds, rash, vomiting. Data Mining is one among the foremost encouraging areas of analysis with the aim of finding helpful information from voluminous knowledge of datasets. Data Mining is especially helpful in medical field when no handiness of proof favouring a treatment is found. Hence the Chronic Kidney Disease (CKD) dataset taken from UCI repository is analysed using different data mining classification. It is observed that LR, SVM techniques have given an accuracy of 93.8% and 94.0% respectively. The techniques RF, KNN, GNB and Voting Classifier have given an accuracy of 100% respectively. For this project, the CKD dataset has been taken from the UCI repository. Classification techniques like Logistic Regression, Support Vector Machine (SVM), K- nearest neighbours (KNN), Random Forest, GNB and Voting Classifier are used in the project.