Preference prediction method and device for consumption by using payment card, equipment, medium and product

Abstract

The invention provides a preference prediction method and device for consumption by using a payment card, equipment, a medium and a product, which can be applied to the technical field of big data and financial science and technology. The method comprises: in response to a preference prediction instruction, obtaining user target features of a plurality of target types of a target user within a first preset time period, the target types being determined from an initial training feature set by using a gradient boosting decision-making tree model in a process of training a target gradient boosting decision-making tree model, the initial training feature set represents asset transaction information of the training user; constructing a target data wide table of the target user according to the multiple user target features; and processing the target data wide table by using the target gradient boosting decision tree model to obtain prediction probability information which represents the probability that the target user consumes by using the payment card.

Images