Network program aggregation and recommendation system and network program aggregation and recommendation method

Abstract

The invention discloses a network program aggregation and recommendation system and a network program aggregation and recommendation method. The network program aggregation and recommendation method comprises the following steps of: constructing a network program source based on network program information in a network program server; extracting program characteristic information showing characteristics of programs from a plurality of programs from a local program source and the network program source; establishing a comprehensive preference model capable of displaying manual preference input information, explicit preference feedback information and implicit preference information of a user for the user; and showing a recommended program list with the program characteristic information to the user based on the comprehensive preference model. According to the method and the system, the program source can be expanded, the user can conveniently indentify the programs in which the user is interested, the favorite programs can be preferentially presented, the information selection is more humanized, and the intelligentization and an interaction function are further improved.

Description

technical field [0001] The invention relates to the field of network multimedia, in particular to a network program aggregation and program recommendation system and method. Background technique [0002] With the vigorous development of my country's radio and television industry, the multi-level program broadcasting system of the central, provincial, prefectural and county levels has been gradually improved, and the number of channels is increasing. Under the current coding and modulation system, the transmission capacity of the cable digital TV system can reach 500-600 programs. Although this meets the viewing needs of various users, it also makes it difficult for users to find the programs they really like among so many programs. At the same time, users hope to see more of their favorite programs, but the current source of programs is very single. Therefore, how to obtain rich program sources is also a problem that needs to be considered. In view of the above, the establi...

AI Technical Summary

Problems solved by technology

[0012] 2. For explicit modeling, there is a big problem, that is, "cold start". Specifically, it means that for users who have just registered, there is no or too little scoring feedback, so it is impossible to draw a shaped preference model from it. At the beginning of the model, there will be obvious recommendation error bias

[0013] 3. For implicit modeling, only the inference of whether the user is interested in the corresponding information can be obtained through the user's browsing behavior, but the degree of user interest cannot be obtained

[0015] 1. Weak initiative

Since the current recommendation technology is mainly content-based search and category browsing technology, users must enter keywords or manually search by category to obtain relevant information, and cannot provide users with active recommendations based on user behavior.

[0016] 2. Single program source

Due to the limitation of recommendation technology, the program source used in TV recommendation is single at present, which cannot bring the richest content experience to users

[0017] 3. Lack of personalization

[0018] 4. It is impossible for users to understand the recommendation results clearly

Because the personalized recommendation system is embedded in the search engine, it is difficult for users to judge which ones are recommended because of their own personalized preferences, and which ones are programs with higher click-through rates on the Internet.

If users cannot understand the feedback results well, they will not be able to give better feedback on their own wishes

Images