Application of an Improved C4.5 Algorithm in Shopping Websites

Decision tree classification is an effective instance-based learning and data-mining method, which is applied to process the large continuous data to save the processing time. In this paper, on the one hand, a K-equal separation method is presented, which reduces the calculation amount and accelerates the tree building speed. On the other hand, the calculation formula of the information gain ratio in C4.5 algorithm is simplified. The improved C4.5 algorithm is applied to analyze users' purchasing behaviors, including helping recommend them the products that they are interested in, and accelerating them to quickly find those goods they want. Experimental results show that the advanced method can effectively improve the training speed of the model.