Analysis of cough detection index based on decision tree and support vector machine

Abstract

In clinical medicine, cough is a common disease. Years of cough diagnosis have been collected from a large number of patients as test data. Using these test data, it is possible to find the hidden rules inside these data, which can improve the diagnosis accuracy of cough. In recent years, these related problems have been concerned by the relevant medical staff. From the known medical data, medical data mining and processing can extract knowledge, and summarize the experiences of medical experts. This technology is becoming more and more important in the medical information field. In this paper, cough test attributes ,such as peak expiratory flow and fractional exhaled nitric oxide (FENO), are modeled by decision tree and support vector machine. The experimental results show that FENO and percentage of eosinophils have a great effect on the diagnosis of cough, which are important attributes for cough diagnosis.