Movie recommendation method and system based on improved deep belief network

Abstract

The invention discloses a movie recommendation method and system based on an improved deep belief network, and the method comprises the steps: obtaining the weight of each movie according to the moviecomment record of a target user, the total number of comments of each movie and the number of users with the same comment record as the target user, and constructing a comment feature vector of the target user; calculating comment similarity according to the comment feature vector of the target user, and constructing a similar user set for the users whose comment similarity is higher than a threshold; constructing a deep belief network by adopting a knowledge migration method, and optimizing the deep belief network model based on a partial least square method to obtain a trained deep belief network; and inputting the movie comment records of the target user and the similar user set into the trained deep belief network, and outputting a movie recommendation result. The recommendation system is combined with deep learning, personalized movie recommendation is made for the target user by efficiently extracting historical comment data of the user on the movie, and the recommendation quality of the recommendation system is improved.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a method and system for movie recommendation based on an improved deep belief network. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] At present, recommendation systems have been widely used on e-commerce platforms. Recommendation systems can provide platform users with personalized recommendations, improve users' experience of the platform, and help companies better recommend products to gain greater benefits. Therefore, the more More and more e-commerce websites use recommendation systems, such as Amazon's product recommendation, Douban's music recommendation, news recommendation, etc. Through the recommendation system, it can not only bring profits to the enterprise, but also satisfy the interests of users. [0004] However,...

AI Technical Summary

Problems solved by technology

[0009] With the in-depth research on neural networks, many methods have been proposed to improve the neural network model, such as determining the artificial neural network structure through the self-organizing structure, but when the self-organizing structure When the organizational structure is applied to DBN, the self-organizing structure will become very complicated, which greatly increases the computational burden; at the same time, the network parameters of the self-organizing structure need to be initialized repeatedly, and the prior knowledge learned before is not fully utilized. Disadvantages of structure-determined neural network structures

Images