A C4.5 decision tree algorithm-based specific user mining system and a method therefor, relating to the field of live website data mining. The method comprises: selecting user samples (S1), wherein the user samples are divided into payment user samples and non-payment user samples according to a classification label; collecting statistics on attribute values of classification attributes of the user samples (S2); performing normalization processing on the attribute values of the classification attributes as training sample data (S3); an algorithm platform receiving the training sample data, and training a C4.5 decision tree model (S4); and after the training of the C4.5 decision tree model is complete, performing normalization processing on new attribute values of the classification attributes of users, and inputting same into the trained C4.5 decision tree model (S5). Under the circumstances of multiple data dimensions and a large amount of data, a C4.5 decision tree algorithm-based specific user mining system can collect comprehensive statistics on indexes related to user interest degree, thereby facilitating mining of a specific user.
展开▼