Intelligent recommendation method and device, computer equipment and storage medium

Abstract

The embodiment of the invention belongs to the technical field of intelligent recommendation of big data, and relates to an intelligent recommendation method, which comprises the steps of obtaining atleast one feature value and an original representation vector of a project, the project comprising a plurality of click sequence projects and a to-be-learned exposure project; splicing the at least one feature value and the original representation vector to obtain a new representation vector of each click sequence item and the exposure item; combining the click sequence items with the new representation vectors of the exposure items, and performing convolution deep learning to obtain convolution features; splicing the convolution features with the processing result of the deep network of theoriginal representation vector, and inputting the spliced convolution features and the processing result into an output layer of a recommendation model to obtain the click probability of an exposure item; and determining whether to recommend the exposure item or not according to the click probability. The invention further provides an intelligent recommendation device, computer equipment and a storage medium. According to the invention, the recommendation accuracy of the recommendation model can be improved.

Description

technical field [0001] The present application relates to the technical field of intelligent recommendation of big data, in particular to an intelligent recommendation method, device, computer equipment and storage medium. Background technique [0002] With the development of network technology, personalized recommendation has become one of the indispensable network services in people's network life, and has also become the focus of the future development of Internet products. Existing technologies usually use deep learning network models for personalized product recommendations. For example, e-commerce companies use deep learning network models when recommending products, generally using wide&deep (width&depth) recommendation models, and the deep network part inputs commodity items (items) The embedding (representation vector), including the click sequence and the representation vector of the exposed product, is connected to the MLP (multi-layer perceptron), and the wide ne...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiment of this application is to propose an intelligent recommendation method, device, computer equipment and storage medium to solve the problem of using deep learning network recommendation results and actual The case does not match the problem

Images