Data processing method and device, computer equipment and storage medium

Abstract

The invention discloses a data processing method which comprises the steps that a server distributes a preset model to each computer device, and the computer devices receive the preset model sent from the server and train the preset model to obtain local parameter data. The computer equipment sends the local parameter data to the server, the server determines updated parameter data corresponding to each piece of local parameter data according to the parameter reward and punishment rule and sends each piece of updated parameter data to the corresponding computer equipment, and the computer equipment carries out iterative training on a preset model according to the updated parameter data to obtain a parameter reward and punishment result; and obtaining a target prediction model until the preset model meets a preset condition, and performing intention prediction on the user based on the target prediction model to obtain an intention prediction result. According to the method, a plurality of participants can be combined to jointly train the intention prediction model, the prediction accuracy of an enterprise for the user with the complaint intention is improved, and the frequency of user complaint is reduced.

Description

Technical field [0001] The present application relates to the technical field of data processing, and more particularly to a data processing method, device, computer device, and storage medium. Background technique [0002] The rapid development of the mobile Internet and intelligent terminals makes the network application product have endless, with product diversity and user base, service providers have grown sharply in the process of providing business services. Since complaint processing is especially important in service providers, it is also an important way to find problems and improve service quality, so reduce user complaints, enhance user satisfaction has become an aspect of service providers' key attention. [0003] In order to be able to predict the complaints that will appear in time, at this stage, most companies mainly use the probability of probability of probability of complaints for users to predict the probability of probabilistic probability of complaints. For ...

AI Technical Summary

Problems solved by technology

However, in practical application scenarios, due to the limitation of training data, service providers can only model based on their own data, resulting in low prediction accuracy and insufficient generalization ability of the model

Images