18-08-2023 дата публикации
Номер: CN116611884A
Принадлежит:
The invention provides a product recommendation method and system based on a multi-dimensional heterogeneous graph neural network in the technical field of product recommendation, and the method comprises the following steps: S10, obtaining user behavior data, user tag data, user social data and product attribute data, fusing the user behavior data, the user tag data, the user social data and the product attribute data to construct a heterogeneous graph; step S20, sampling neighbor nodes of each node in the heterogeneous graph to obtain related neighbor nodes; step S30, performing double-layer aggregation on each node in the heterogeneous graph based on the related neighbor nodes to obtain a fusion node; step S40, creating a product recommendation model based on the neural network, and training the product recommendation model by using the fusion node; and S50, carrying out product recommendation by utilizing the trained product recommendation model. The method has the advantage that the ...
Подробнее