Movie recommendation system based on cross-modal fusion

Abstract

A movie recommendation system based on cross-modal fusion comprises an input module, where the input information comprises feature information of each user in a user set, feature information of each movie in a movie set, interaction information of each user in the user set to each movie in the movie set, and poster pictures of each movie in the movie set; the system also comprises a preprocessing module, a single-mode coding module, a cross-mode fusion module and an output module. Visual information of movie poster pictures is extracted based on a deep neural network model, and useful information which can be used in the recommendation process is enriched; on the other hand, interaction and fusion of text and visual information are realized based on a cross-modal fusion algorithm, so that movie recommendation is better carried out on the user; compared with a traditional recommendation algorithm which only uses single-mode information, does not interact two-mode feature information and does not distinguish single-mode and multi-mode interaction modes, the invention has a better recommendation effect.

Description

technical field [0001] The present invention relates to information recommendation technology, more specifically, to a movie recommendation system based on cross-modal fusion. Background technique [0002] With the explosive growth of network information, people will face a dazzling array of options when making movie choices, and the information people receive is much larger than the information they actually want, which causes the problem of information overload. The recommendation system can alleviate this problem. It can filter information according to the user's individual needs so that the movies presented in front of the user are what the user really needs. In the field of movie recommendation, the number of users and movies is very large, which will lead to very scarce data on the actual interaction of users with movies, and this interaction data is the key data used by the recommendation algorithm, so this data is sparse Sexuality is an important issue that limits t...

AI Technical Summary

Problems solved by technology

[0004] Moreover, considering that the work of using poster images as visual information for movie recommendation is mainly to extract features from text information and visual information and then directly splicing them together and sending them to some fusion modules for information fusion, it has not been able to achieve the two models. Interactive operation of state feature information, such as the VBPR algorithm model (refer to the paper He R, McauleyJ. VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback[J].2015.)

[0005] In addition, the splicing and fusion of two different modal features cannot simply be equivalent to a new single-modal feature information to perform the same interactive operation

However, the existing multimodal recommendation algorithms do not distinguish between single-modal and multimodal feature processing, such as the DICM algorithm model (refer to the paper Tiezheng Ge, Liqin Zhao, Guoorui Zhou, etal. Image Matters: Visually modeling user behaviors using Advanced Model Server[C] / / The 27th ACM International Conference on Information and Knowledge Management, 2018.)

Images