A social image recommendation method based on hierarchical attention mechanism

Abstract

The invention discloses a social image recommendation method based on the hierarchical attention mechanism, which comprises the following steps: 1. Constructing heterogeneous data: a scoring matrix ofan image by a user, an uploading information matrix of the image by the user, and a social relationship matrix between the user and the user; 2. Processing heterogeneous data and image set to obtainsocial embedding matrix, content embedding matrix and style embedding matrix; 3. Input that embedding matrix into the bottom attention network to obtain the social semantic information; 4. The weightsof the three aspects of the social semantic information obtained from the top-level attention network when the user chooses the image; 5. According to the feature matrix, three kinds of social semantic information matrices and their weights, we can get the scoring prediction value, and then we can recommend the image. The invention not only effectively solves the problem of data sparsity by usingthe uploading information of the user and the social information among the users, but also well explains the user preference through the hierarchical attention mechanism, and realizes the accurate social image recommendation.

Description

technical field [0001] The invention relates to the field of image recommendation, in particular to a hierarchical attention-based social image recommendation method. technical background [0002] Image-based social networks have become the most popular social networks in recent years. With the rapid increase of mobile phone users, many users take pictures and upload them to social networking platforms to share their lives. A large number of uploaded images lead to image overload. How to understand the preferences of different users and make accurate image recommendations has become an urgent need. [0003] Collaborative filtering algorithm can effectively solve the problem of image recommendation. It discovers user preferences and makes recommendations by mining user historical behavior data. Although the collaborative filtering algorithm is widely used, the sparseness of the user-image interaction behavior matrix limits its recommendation performance. In order to solve ...

AI Technical Summary

Problems solved by technology

[0004] However, the collaborative filtering method based on image visual features only considers the intrinsic attributes of items and does not take advantage of the intrinsic attributes of users.

Images