Multi-view attention recommendation algorithm based on binary information network

Abstract

The invention provides a multi-view attention recommendation algorithm based on a binary information network. According to the algorithm, a plurality of high-quality paths from a target user to targetcommodities are generated from the binary information network; CNN (Convolutional Neural Network) and max-pooling operations are adopted for the generated path to extract a corresponding path vector.Weighted merging is performed on generated multiple path vectors through an attention mechanism to obtain a path merging vector capable of corresponding to a target user and a target commodity pair.Similarly, the user vector and the commodity vector are updated by utilizing the corresponding path merging vector generated in the S3 through the action operation; and the generated path merging vector and the user vector and the commodity vector are spliced, and the spliced vectors are transmitted to a multi-layer perceptron for training to obtain final scoring prediction.

Description

technical field [0001] The present invention relates to the field of graph neural networks, more specifically, to a multi-view attention recommendation algorithm based on binary information networks. Background technique [0002] In recent years, with the vigorous development of the Internet economy, recommendation algorithms have been applied to all aspects of people's lives. How to efficiently realize user-oriented personalized recommendations has become an important research direction for many companies. Among the commonly used recommendation algorithms, there is a method that is applied to most scenarios, that is, the collaborative filtering algorithm, which can be divided into content-based collaborative filtering (such as user or product-based KNN algorithm) and simulated interactive behavior-based Collaborative filtering (such as collaborative filtering based on matrix decomposition), the latter effect is particularly obvious, while it has received a lot of attention,...

AI Technical Summary

Problems solved by technology

Such information can be understood as the in-depth information contained in the interaction between users and products. They are obviously ubiquitous, and the collaborative filtering algorithm mentioned above cannot take this information into consideration.

Images