Product information recommendation method and device

Abstract

The invention provides a product information recommendation method and device. The method comprises the steps of receiving customer basic information of a target customer; determining preference scores of the target customer corresponding to different product types by applying the customer basic information and a plurality of preset product preference prediction models corresponding to the different product types, and taking the product type corresponding to the maximum preference score as a target recommended product type; sending product purchase information corresponding to the target recommended product type to the target customer, wherein each preset product preference prediction model is an iterative decision tree model obtained by pre-training according to the historical customer basic information of a plurality of historical customers and the respective preference score label of each historical customer for the same product type. According to the invention, the timeliness and intelligence degree of product recommendation can be improved, and the success rate of product recommendation can be further improved.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a method and device for recommending product information. Background technique [0002] As bank wealth management products become more diverse and complex, and product updates are faster and faster, how to quickly understand customer needs and timely grasp changes in customer demands under the new situation is a new opportunity and challenge for banks' wealth management business . At the same time, banks face a large customer base, difficult to discover customer needs, and high maintenance costs. Due to limited human resources, most small and medium-sized enterprise customers are in a state of lack of management and marketing, and can only use technical means to manage long-tail customer marketing. . [0003] Using traditional big data analysis or machine learning methods for product recommendation, the model relies on massive historical data for training, and...

AI Technical Summary

Problems solved by technology

This method has contradictions in the product recommendation under the new situation, that is, adopting a large amount of historical data for training. Although the model fits the historical laws well, it cannot take into account the new situation changes and customers' preferences for new products. The laws obtained by the model may be "too old"; if only recent data are used for training, the amount of training data will be greatly reduced, the overall fitting effect of the model will decrease, and the laws obtained by the model may be "too few"

Model training is caught in a dilemma, unable to guarantee the final product recommendation effect, and it is difficult to meet the requirements of the new situation of wealth management sales

Images