Movie recommendation method and apparatus

Abstract

The invention is suitable for the technical field of recommendation algorithm applications and provides a movie recommendation method and apparatus. The method comprises the steps of identifying and judging whether a target user is a registered user or not; if the target user is the registered user, judging whether an evaluation number of the target user for a movie is greater than or equal to a preset evaluation threshold; if the evaluation number is greater than or equal to the preset evaluation threshold, recommending the movie to the target user by a movie feature-based collaborative filtering recommendation model; and if the target user is not the registered user or the evaluation data is smaller than the preset evaluation threshold, recommending the movie to the target user by a WEB usage mining-based recommendation model. According to the movie recommendation method, corresponding recommendation methods can be used for different target users, so that the movies recommended to the target users can conform to the interests and preferences of the target users to the maximum extent; and the problem that the movies cannot be recommended to new users or target users with relatively small movie evaluation numbers in a collaborative filtering-based single recommendation algorithm in the prior art can be solved.

Description

technical field [0001] The invention belongs to the technical field of recommendation algorithm application, and in particular relates to a movie recommendation method and device. Background technique [0002] The recommendation system has a good development and application prospect in the electronic commerce system, and gradually becomes an important research content of electronic commerce IT technology. The biggest advantage of the recommendation system is to collect the information that the user is interested in, analyze the information to obtain the user's interest preference, and make personalized recommendations for the user according to the user's interest preference. At present, recommender systems have been widely used in different fields, such as movie recommendation. The recommendation algorithm based on collaborative filtering is favored because it is easy to tap the potential new interests of users. The current movie recommendation system basically uses this si...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a movie recommendation method and device to solve the problem in the prior art that a single recommendation algorithm based on collaborative filtering cannot recommend movies to new users or target users with fewer movie reviews

Images