Customized information pushing method and device and electronic equipment

Abstract

The invention provides a customized information pushing method, which is used for pushing customized information to a user, and comprises the following steps: when an access request of a current useris received, obtaining user characteristic data of the user; performing target user screening according to the user feature data; constructing a real-time classification model, training the real-timeclassification model by using training data, the training data including user attribute data and financial performance data, and the financial performance data including credit granting data and dynamic branch data; calculating financial evaluation values of the screened target users by using a trained real-time classification model, and classifying the target users in real time; and pushing customized information to the user according to the financial prediction value of the target user. According to the method, the conversion rate is improved, the information pushing cost is effectively reduced, and the accuracy of customized information pushing is improved.

Description

technical field [0001] The invention relates to the field of computer information processing, in particular to a custom information push method, device and electronic equipment. Background technique [0002] With the vigorous development of mobile communications, telecom operators can process and store more and more telecom data. In order to efficiently find target users, many companies and enterprises have begun to cooperate with telecom operators to obtain their telecom data for analysis and analysis. After the results are obtained, they will be fed back to the cooperative merchants for accurate information push. [0003] In the prior art, many telecommunication companies have widely used data mining in customer segmentation, and some telecommunication companies attach great importance to customer segmentation and customer information push, and make correct decisions based on the results of customer segmentation. For example, a telecom company has a wide range of applicat...

AI Technical Summary

Problems solved by technology

[0004] Although big data has some applications in precise information push, there are still few precise information push methods for telecom big data. The main reason is that telecom operators will not casually disclose users’ telecom data in order to protect user privacy.

However, there are still the following problems in information push: 1) The efficiency of information push is low. In the period when major operators focus on internal business, most of the products pushed by SMS precise information are voice, call, and traffic products. According to business experience Design information push customer group extraction caliber, but cannot meet the information push needs of a large number of multi-category products in the same period

2) The effect of information push is difficult to guarantee. With the increasing diversification of information push products, it is almost impossible to expand the business knowledge reserve of business experts simultaneously. The method of relying solely on information push personnel to design information push caliber cannot satisfy high-quality customer groups Acquisition is limited by the subjective cognition level of the business, and it is impossible to push personalized and accurate information

Images