Application of Decision Tree Classification Method Based on Information Entropy to Web Marketing

The fast and accurate customer classifier is the core of Web marketing. But the common customer classifier doesn´t fit for the customer classification in Web marketing, because it has certain industry limitation. In this paper, a decision tree classifier based on information entropy is proposed. First of all, the classification index system for Web marketing is established by integrating several common-used classification index strategies. After that, the information gain of each classification index is calculated by using information entropy. Finally, the decision tree is established according to the information gain of classification index, and the corresponding customer classification rule is generated. To test this method, the real e-commerce site data is divided into two parts. One is used to establish the customer classification decision tree. The other is used to test. The experiment result shows that the forecast accuracy of decision tree is about 97%, which meets the requirement of the actual classification work.