Study on classification rules of hypertension based on decision tree

To extract classification rules of hypertension more effectively, the C4.5 decision tree algorithm was put forward. Firstly, C4.5 algorithm was used to analyse pre-processed hypertension data and a decision tree model was constructed. Then, the understandable classification rules were extracted and the maintaining method was used to test the accuracy of the classification results. Finally, the improved C4.5 algorithm was proposed to amend the information gain of the selected attribute by introducing the concept of correlation. The improved decision tree model more conforms to the medical understanding and the classification accuracy was also improved. The experiment results show that the improved algorithm was effective.