A Medical Expert System to Manage Disease using Oversampling and Decision Tree

Abstract

Now-a-days modern technology is joining hands with traditional system of medical diagnosis and treatment to alleviate health related problems faced by modern age due to its lifestyle and hectic work pressure. Therefore, it is paramount to develop a system that can identify the major causes of health related problems beforehand, so that people can follow some precautionary measures to prevent those causes and can have hassle free life. Decision tree is one of the reliable and popular data mining tools to generate production rules to identify significant causes for a disease. However, imbalanced nature of the dataset makes it difficult for the decision tree to learn for the minority class. Therefore, this paper proposes a model called Balanced Medical Expert System for Managing Disease (BMESMD) using an oversampling technique with decision tree that identifies major cause (s) with ranges of the disease. Ranges of the major cause (s) may be controlled by medicine, food habit or exercise prescribed by the doctor. The proposed method is validated with three medical datasets taken from UCI repository. It is observed from the experimental results that the proposed model can manage the disease significantly by controlling one or two major causes.

Keywords: Medical Expert System; Machine learning; Decision tree; Imbalanced problem; oversampling

Cite this Article: Prabhudatta Kar, Trisha Bhattacharyya, Disha Mohanty, Sai Sachin Vutukuri, Manomita Chakraborty, Saroj Kr. Biswas. A Medical Expert System to Manage Disease using Oversampling and Decision Tree. Recent Trends in Programming Languages. 2020; 7(1): 27–33p.