P2P (Peer-to-Peer) propagation type content recommendation method

Abstract

The invention relates to a P2P propagation type content recommendation method. The method comprises the following steps: a user U1 receives content push which is from a system or interested by a highly accepted user U0 of U1; the system judges whether the U1 is interested in the push content or not according to the reaction of the user U0 to the push content and a set standard; if the judgment result is that the user U1 is interested in the pushed content, the recognition degree of the U1 to the U0 is increased, and the same content is pushed to other users who do not push the content and are highly recognized to the U1; and if the judgment result is that the U1 is not interested in the pushed content, the recognition degree of the U1 on the U0 is reduced. The invention provides the P2P (Peer-to-Peer) propagation type content recommendation method by imitating a scene of recommending an article worthy of recommendation to other people with good relationship when people communicate with one another in real life, and solves the problem that the recommendation range of the combined recommendation based on collaborative filtering and content is limited by similar users and interest points and the problem of high operand. And the surprise degree of the user is improved.

Description

technical field [0001] The invention belongs to the field of Internet multimedia, and in particular relates to a recommendation method in the field of information push, commodity push, and Internet multimedia, especially a P2P transmission-type content recommendation method. Background technique [0002] The recommendation system usually determines the user's interests based on the user's browsing habits, and recommends appropriate information to the user by mining the user's behavior to meet the individual needs of the user and help the user find interesting but difficult-to-find information or products. [0003] The recommendation system has a large number of applications in both the Internet and traditional industries. In the Internet industry, almost all Internet platforms have applied recommendation systems, such as content recommendations for news / movies / knowledge communities, product recommendations for e-commerce platforms, etc.; in traditional industries, some are u...

AI Technical Summary

Problems solved by technology

While the recommended content is gradually accurate, the surprise is too low, and the scope of push is gradually limited, which has become the main problem

[0005] ZL2018107716171 discloses a hybrid recommendation method that incorporates content information. This method obtains a distributed dense vector representation from the content text and combines it with the scoring data, and incorporates the noise reduction autoencoder in deep learning technology, focusing on the content from the scoring matrix. Learning hidden feature representation to reconstruct the scoring matrix for prediction, but the recommendation of this recommendation method still uses the scoring matrix and scoring data to predict, and the accuracy of the recommended content is not high

[0006] ZL2016109093975 discloses an information diffusion method and system for accurate promotion and dissemination of information. By combining the user's mobile terminal to send an association invitation request to increase the number of user terminals and the server to recommend the user terminal, it can play a good role in motivating the user's mobile terminal. The role of independently expanding the number of platform users is helpful to quickly expand and spread the number of platform users, so that information can be disseminated quickly and effectively. Although this method achieves the diffusion of information, the scope of its push is very limited

Images