Business opportunity recommendation method and system based on CNN model, electronic equipment and medium

Abstract

The embodiment of the invention discloses a business opportunity recommendation method based on a CNN model. The method comprises the steps of obtaining to-be-processed data; preprocessing the to-be-processed data to obtain CNN, CFR and XGB model data sets; adopting a CNN model to predict and obtain a target user; adopting a CFR model to predict and obtain a business opportunity preliminary sequence in which each target user is interested; for each target user, calling the corresponding XGB model data set, and inputting the XGB model data set and the business opportunity preliminary sorting into the XGB model to obtain the business opportunity final sorting of each target user; and finally sorting and distributing the user IDs and the business opportunities to different business departments. Compared with an existing manual business opportunity pushing method, the business opportunity recommendation method has the advantages that business opportunity pushing can be completed within twohours, the coverage of users is deeper and more comprehensive, new and old users can be well mined, the daily pushed business opportunities and user conversion rate are high, and a good assistance effect is achieved for business departments.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a CNN model-based business opportunity recommendation method, system, electronic equipment and media. Background technique [0002] Service platform companies artificially select target customers from massive databases, and then push them to sales staff based on existing experience and business logic. Next, push the most likely business opportunities for users. [0003] There are two main problems in the existing business opportunity recommendation methods: [0004] 1. User mining: mainly business personnel use business logic and experience to filter recent data, which is time-consuming and labor-intensive, and the effect of user screening is not necessarily good, and most potential target users will be lost. [0005] 2. Forecasting business opportunities: It is mainly recommended by human resources through certain rules and experience. The type of recommended bus...

AI Technical Summary

Problems solved by technology

[0005] 2. Forecasting business opportunities: It is mainly recommended by human resources through certain rules and experience. The type of recommended business opportunities mainly depends on the experience reserve of business personnel, and the accuracy of prediction is low, resulting in a low performance conversion rate of sales personnel

Images