Data processing method and device for information pushing

Abstract

The invention relates to a data processing method and device for information pushing. The method comprises the steps of using a determining module for determining a target user needing information pushing; acquiring first historical record information corresponding to the target user by using a record acquisition module; determining a user type and user characteristics of the target user accordingto the first historical record information by using a user information determination module; acquiring candidate push information corresponding to the target user according to the user characteristics by using an information acquisition module; obtaining a pre-trained prediction model corresponding to the user type by using a model obtaining module; and obtaining a predicted recommendation resultcorresponding to the candidate push information by using a result obtaining module. According to the method, during recommendation, all the feature information of the user can be combined instead ofbeing sorted according to one attribute, so that a more accurate recommendation result can be obtained, and the information click rate and the order forming rate of the pushed user are increased; meanwhile, grouping sorting is not needed, the recommendation efficiency can be greatly improved, and the manpower and material resource cost is saved.

Description

technical field [0001] The present application relates to the field of Internet technologies, and in particular to a data processing method and device for information push. Background technique [0002] Activity recommendation, that is, to display activities related to products or content in different resource locations for users who browse the APP, so as to achieve a certain purpose of operation transformation. [0003] At present, the display of activities requires the configuration and adjustment of special operation personnel. After the front-end users are divided into different types according to the priority, the display is artificially matched and displayed during the event, and different activities are displayed to users of different layers. This traditional method has the advantages of Strongly subjective and empirical. When large-scale events need to be held, many users are involved, product complexity, and event diversity are involved. At this time, a lot of manp...

AI Technical Summary

Problems solved by technology

When the iteration of the activity is replaced, the above process must be repeated, resulting in low efficiency of the activity display iteration and easy display errors due to logical loopholes

[0004] In addition, although event recommendation has some relatively mature development in the advertising exposure process and some APP advertising positions, in the non-advertising field and some scenarios that need to combine the user's financial and wealth attributes, most of the event recommendation related technologies still adopt pre-advertising User grouping, artificially sorted according to priority, and displayed on the front end of the resource slot, there is no technology for automatic recommendation of activities

[0005] It can be seen from the above that in the prior art, the use of front-end user grouping for sorting requires a lot of manpower, so the efficiency is low; in addition, users are only grouped by different attribute information of users, so it is impossible to push advertisements in this way. Taking into account all the characteristics of the user, resulting in low targeting, which further leads to the problem of low click-through rate of advertisements

[0006] Aiming at many technical problems in related technologies, no effective solutions have been provided so far

Images