Diagnosis of Chronic Kidney Disease Using Random Forest Classification Technique

Author Name(s): *Tilakachuri Balakrishna, B.Narendra, Mooray Harika Reddy, Damarapati Jayasri
Author Email: balakrishna.tilakachuri@yahoo.co.in

Abstract

Diagnosing a disease is a complicated task in many existing medical expert systems, Diagnosing a disease is based on the patient symptoms and other details that are given as input to the system. Several levels of uncertainty are involved in medical diagnosis. Data mining is a dominant research area to diagnose the medical diseases. Random Forest is one of the classification technique used to diagnose chronic kidney disease. UCI is a Machine learning Repository which maintains a large collection of medical datasets. Chronic kidney disease data set is one of the dataset used for the proposed model. The proposed work shows the Random Forest Classification is best suitable classifier for Chronic Kidney Disease Dataset among all classifiers like Naive Bayes, SMO, J48 et.c., With these proposed method an Diagnosis system is developed to produce results to end users.

Keywords

Data Mining, Classification, Random Forest Classification.

Introduction

Data mining is an intelligent process to extract useful information in the stored data, now a days data mining techniques are popularly used to diagnosing diseases in the medical field [1]. Classification[2] is one of the prominent area in data mining also known as supervised learning which refer to categorization is a the process in which ideas and objects are recognized, differentiated, and understood . It predicts categorical class labels and classifies data (constructs a model) based on the training set and the values (class labels) to classify new data.

Conclusion

Data mining techniques are used to predict the chronic kidney disease. Based on the percentage of correctly classified instances, we found that the random forest classification technique is best suitable compared to other classification techniques like J48,NB Tree (Naïve Bayes) and REP Tree and so on. The disease diagnosis can also be found in the patients by using android apps.

158 total views, 2 views today

Download PDF File

About the author: admin