Research on Hepatitis Auxiliary Diagnosis Based on Random Forest and Support Vector Machine

Abstract

Hepatitis not only endangers the health and life of patients, but also causes a heavy burden for their family on the society. It has become an important disease with serious social and public health problems. In this study, we studied a large sample of people and obtained biochemical data of patients related to the hospital, including a series of continuous and discrete data, such as age, bilirubin, alk_phosphate, Sgot, Albumin, etc. Then support vector machine (SVM) and random forest model were constructed to assist hepatitis. The SVM with RBF kernel is the best experimental model, which has good performance in the evaluation of accuracy and ROC. Next, we can provide reference and help for clinicians to make clinical decisions based on the results of the experiment, so as to improve the diagnostic accuracy of the non-invasive diagnosis.