Symptom selection of inquiry diagnosis data for coronary heart disease in Traditional Chinese Medicine by using social network techniques

Coronary heart disease (CHD) is a common cardiovascular disease in the elderly, which causes high death rate and low cure rate. Therefore, the TCM diagnosis objectification is important. This paper proposes the Relative Associated Density (RAD) method to analyze the data set. RAD results of the symptoms to syndromes are used in performing feature selection, and the prediction results with different classification machines show significant improvements. Compared to other traditional feature selection methods, RAD provides higher Interpretability in the TCM field.