Business prediction method and device, equipment, storage medium and program product

Abstract

The invention discloses a service prediction method, device and equipment, a storage medium and a program product, and the method comprises the steps: determining a first federated participant and each second federated participant corresponding to a federated model identifier in a service prediction request when the service prediction request which is sent by a user side and carries a service parameter is received; acquiring sample information from the service prediction request, and generating a model prediction request carrying service parameters required by prediction; and sending the model prediction request to the first federated participant, so that the first federated participant generates a corresponding federated prediction result in combination with each second federated participant based on the federated model prediction request. According to the invention, only the request with the service parameter needs to be provided, the request can be forwarded to each federation participant through the server to obtain the federation prediction result, the defect of data privacy leakage in the prior art is overcome, and the privacy security problem is solved from the source.

Description

technical field [0001] The present invention relates to the technical field of machine learning computing, and in particular, to a business prediction method, apparatus, device, storage medium and program product. Background technique [0002] Before users call the finished model service through the API-based online interface, they need to evaluate whether the finished model service is helpful for business improvement, and then decide whether to use the finished model service online. At present, when the user evaluates the finished model service, the user will The predicted sample needs to be handed over to the provider of the finished model service, so that the provider can find the finished federated model service corresponding to the finished model service, and return the prediction result of the predicted sample based on the finished federated model to the user. In this process, even if The prediction samples handed by the user to the finished model service provider are ...

AI Technical Summary

Problems solved by technology

[0002] Before calling the finished model service through the API-based online interface, the user needs to evaluate whether the finished model service is helpful to business improvement, and then decide whether to call and access it online. Currently, when the user evaluates the finished model service, the user will The forecast samples need to be handed over to the provider of the finished model service, so that the provider can find the federated model product corresponding to the finished model service, and return the prediction result of the forecast sample based on the federated model finished product to the user. In the process, even The prediction samples submitted by the user to the provider of the finished model service have been encrypted, and there is still a risk of data privacy leakage

Images