The Description of Optimal Decision Tree Algorithm and Its Application in Customer Consumption Behavior

This paper describes the basic idea and main disadvantages of decision tree algorithm, proposes to add optimization parameters to the information gain algorithm, and weighted important attributes, so that the small number of data tuples will not be ignored in the generation of decision tree, and at the same time, reduce the rule results of decision tree by optimizing pruning, and form more valuable decision information. This algorithm model is applied in the analysis of customers' consumption habits, which improves the accuracy of classification results and prediction information, and provides more targeted decision support for enterprises to make sales strategies.