Method and method for recommending data
A data recommendation and data technology, which is applied in video data retrieval, electronic digital data processing, special data processing applications, etc., can solve the problem of unable to realize personalized recommendation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0037] like figure 1 As shown, the present invention provides a data recommendation method, including:
[0038] Step S11, using the features of all users and the number of the first data to be recommended as feature factors to train the weight factors of all users. Wherein, the data may be a video, and the serial number of the first data to be recommended may be a video ID, and different user characteristics represent different preferences of the data to be recommended.
[0039] Preferably, the characteristics of the user can be obtained from a user behavior log, and the characteristics of the user include browser type, display resolution, network device type, time of visiting the website, location, user website referrer and One or any combination of the user's landing page (landing page), the user's characteristics include one or more characteristic factors, of course, the user's characteristics can also include the data displayed by the user, the data clicked or watched by ...
Embodiment 2
[0065] like figure 2 As shown, the present invention provides another data recommendation method. The difference between this embodiment and the embodiment is that the first data to be recommended is high-quality data obtained regularly, and the high-quality data is classified according to the characteristics of one or more users. Sorting, obtaining the top Q high-quality data as the second recommendation data, so as to make the recommendation result more accurate, the method includes:
[0066] In step S21, the regularly acquired high-quality data is used as the first data to be recommended; specifically, the data can be videos, for example, high-quality videos can be regularly obtained from a video library of the entire network and updated and stored in a high-quality video library. There may be tens of millions or even hundreds of millions of videos in the online video library, and the amount of data is too large. It will be a lot of work to train the weight factors of all ...
Embodiment 3
[0095] like Figure 4 As shown, the present invention also provides a data recommendation system, including a model module 1 and a recommendation engine module 2 . Wherein, the data may be a video, and the serial number of the first data to be recommended may be a video ID.
[0096] The model module 1 is used to use the features of all users and the number of the first data to be recommended as feature factors to train the weight factors of all users.
[0097] Preferably, the number of weighting factors=1+the number of features of all users×the number of data to be recommended.
[0098] Preferably, the characteristics of the user can be obtained from a user behavior log, and the characteristics of the user include browser type, display resolution, network device type, time of visiting the website, location, user website referrer and One or any combination of the user's landing page (landing page), the user's characteristics include one or more characteristic factors, of cour...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com