A Movie Recommendation Method

Abstract

The invention discloses a movie recommendation method and belongs to the technical field of recommendation systems. First of all, it is considered that the same type of movies have similar characteristics in poster design style; therefore, it is necessary to obtain movie-related poster information, and use variational self-encoding to extract features from these movie posters, and at the same time use the extracted poster features as the potential of the movie. Feature vectors, according to the potential feature vectors of the movies, use the cosine similarity to calculate the similarity between movies, then rank them according to the similarity, and finally select the most similar movies as the recommendation results to recommend to users. This method has low complexity and has no shortcomings such as sparseness and cold start in the traditional collaborative filtering recommendation method. Mainly used for movie recommendation.

Description

technical field [0001] The invention belongs to the technical field of recommendation systems. Background technique [0002] With the rapid development of the Internet, the information that people recognize has experienced explosive growth. The information generated in the past thirty years is more than the sum of the information generated by human beings in the past thousands of years. In the case of such a large amount of information, it becomes very difficult for users to obtain effective information. Therefore, in order to overcome the problem of "information overload", the recommendation system came into being. [0003] The movie recommendation system is an important application in today's recommendation system. Traditional movie recommendation systems generally use collaborative filtering algorithms. Interesting information, the main advantage of collaborative filtering algorithm is that it can recommend two unrelated items, but collaborative filtering also faces pro...

AI Technical Summary

Problems solved by technology

[0008] It can be seen that although the existing methods have certain advantages, there are also some shortcomings: the CNN (convolutional neural network) feature is used, and the CNN neural network is a supervised neural network. At the same time, the training samples need to manually mark the data. This is not conducive to processing large-scale data

[0009] At present, there are few methods of applying neural networks to recommend movies, and most of them use supervised neural networks, and there are few methods using unsupervised neural networks; in order to improve the efficiency of training neural networks and get higher movie recommendations Accuracy, in view of the deficiencies of the existing solutions stated above, the present invention aims to provide a more efficient and more complete solution, and overcome the defects in the prior art

Images