期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：4
页码：6650
DOI：10.15680/IJIRCCE.2017.0504011
出版社：S&S Publications
摘要：We present a method to analyse the risk of diseases like diabetes, sexually transmitted infections,increased blood pressure, breast cancer, heart disease and chronic kidney disease based on the patient symptoms, pastdiagnosis and lifestyle. The presented methodology may be incorporated in applications like communication anddecision support systems in health care, risk management, health analysis and disease prevention.We use datasets ofdifferent diseases from sources like data.gov.in, UC Irvine(UCI) machine learning datasets, pimaindian diabetes data,etc. When a user enters the symptoms related to a disease we classify the patient in one of the disease categories. Thentaking the patient lifestyle in account, we analyse the degree of risk for the particular disease.We use Naive Bayesclassifier and C4.5 decision tree to classify patients in various categories. These classifiers can also be compared withother classifiers like logistic regression, artificial neural networks, support vector machines, random forests, baggingand boosting.In case of high dimensional data, it can be reduced using principal component analysis (PCA) and randomsub sampling. The method we proposed will predict accurate analysis of patient data.
