Potential customer identification method and system based on longitudinal federated learning, and medium

Abstract

The invention discloses a potential customer identification method and system based on longitudinal federated learning and a medium, and the method comprises the steps that a first loss value of an updated first preset logistic regression model is obtained through the calculation of a local end, and a second loss value of an updated second preset logistic regression model is obtained through the calculation of a participating end; finally, the local end and the participating end judge whether all preset stopping conditions are met at the same time or not according to the corresponding first loss value and the second loss value, and if yes, the local end and the participating end recognize the customer according to a preset potential customer recognition method so as to complete recognition of potential customers; and if not, the appointed steps are repeatedly executed, and judgment is carried out again. According to the potential customer identification method and system based on longitudinal federated learning and the medium provided by the embodiment of the invention, the accuracy of identifying potential customers is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method, system and medium for identifying potential customers based on longitudinal federated learning. Background technique [0002] At present, with the increasingly high level of technology and digitalization of life, the public has generated a large amount of private data in various aspects such as food, clothing, housing, transportation, and use. These all-round data provide artificial intelligence, blockchain, cloud computing, Innovative applications of emerging technologies such as big data provide fertile ground. At the same time, how to use emerging technologies to safely and efficiently mine user data and create commercial value has increasingly become the focus of financial institutions such as banks. Because customers will generate a large amount of financial transaction data when they handle financial services such as account opening, transfer, deposit a...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a potential customer identification method, system and medium based on vertical federated learning to solve the problem of poor prediction effect of using a single model to identify potential customers in the prior art. By adopting a joint modeling method based on vertical federated learning Identify potential customers to improve the accuracy of identifying potential customers

Images