Operation content generating and pushing method and device

Abstract

The invention provides an operation content generating and pushing method and device, which can be applied to the financial field, and the method comprises the steps: determining a target customer group according to obtained customer data, and generating a customer portrait by using a pre-constructed customer portrait model; according to the target customer group and the customer portrait, generating a decision operation event by using a pre-constructed decision operation generation model; capturing real-time behavior data of the customer, and dynamically generating a real-time operation event according to the real-time behavior data and the customer portrait; and generating operation content according to the decision operation event or the real-time operation event and pushing the operation content. According to the method, customer group screening strategies can be flexibly designed according to different operation requirements, the technical effects of accurately determining operation customer groups and recommending products in a personalized mode are achieved, accurate decision-making of operation events can be achieved, operation real-time performance is improved by generating transaction accompanying operation events in real time, and user experience is improved. And the operation effect backflow can be used for optimizing the model, so that the technical effects of operation refinement, high timeliness, low cost and high efficiency are achieved.

Description

technical field [0001] The present application belongs to the technical field of machine learning, and specifically relates to a method and device for generating and pushing operation content. Background technique [0002] At present, in the field of customer operation in the banking industry, the target customer groups are different for marketing activities with different themes. In the traditional marketing model, in order to reach more customers, an extensive network operation content recommendation model is often adopted, but the investment cost is often high but the effect is low. In the previous operating methods, the method of proactive marketing without background events was generally used to reach customers, which lacked the combination with customer needs or behaviors. After a transaction occurs, users often expect to get discounts or rebates, so as to make up for expenses or obtain greater income. At this time, the timeliness of customer marketing is particularly...

AI Technical Summary

Problems solved by technology

[0004] This application provides a method and device for generating and pushing operation content, to at least solve the problem that many customer operation systems in the current banking industry are still in the stage of manually exporting results and then manually processing the operation results, and the marketing positioning recommendations for customers are not accurate enough and the effect feedback It is not intuitive and concise enough, and it has not been able to return the data to form a closed loop of the system, and the operation result data has not fully exerted its value

Images