Patient Opinion Mining: Analysis of Patient Drugs Satisfaction using Support Vector Machine and Logistic Regression Algorithm

Dina R. Nahma; Ayad R. Abbas

Dina R. Nahma Computer Science Department \ University Of Technology, Baghdad, Iraq
Ayad R. Abbas Computer Science Department \ University Of Technology, Baghdad, Iraq

Keywords: Opinions mining; Support vector machine; Regression algorithm; Drugs reviews; Drugs Satisfaction

Abstract

Patients opinion mining is reflect patients' observations and perspectives on medical drugs or treatments in drug reviews. The Opinion Model used in this paper focuses on predicting the degree of drug satisfaction among patients. This work aims to apply social-web opinion mining machine learning methods to healthcare. This data set was collected from a web said UCI learning software repository with two sets of data: train and test (75-25%). Where divided the drug ranking into three general classes: positive (7-10), negative (1-4) or neutral (4-7).Where examine reviews of different drugs in this research, which were analyzed in text format and also assessed on a scale from 1 to 10. After that input the dataset to classification by two algorithms used the support vector machine and Logistic Regression algorithm, used this algorithm to improve the results of support vector machine algorithm. The support vector method with logistic regression yields better results in terms of accuracy.