A training method and device for an information recommendation system
A technology for recommending information and recommending information, applied in the computer field, can solve the problem that the recommendation system cannot train the sorting model, etc., and achieve the effect of improving accuracy and protecting data privacy
Active Publication Date: 2020-12-18
WEBANK (CHINA)
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0012] The present invention provides a training method and device for an information recommendation system, which solves the problem that the recommendation system cannot train a sorting model and generate new model parameters in a cold start scenario
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0077] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0078] In order to solve the problem that the recommendation system in the prior art cannot train the ranking model and generate new model parameters in the cold start scenario, in the embodiment of the present invention, a loop iteration method is adopted, based on the feedback behavior of the user to each recommendation information in the recommendation information set , and the second click label of other users sent by the server, fir...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a training method and device for an information recommendation system, which solves the problem that the recommendation system cannot train a sorting model and generate new model parameters in a cold start scenario. The feedback behavior of each recommendation information in the information set, as well as the second click label of other users sent by the server, first update the recommendation index parameter set of the ranking model for the first time; The aggregation parameter set of the recommended index parameter set of the recommended index parameter set is updated again until the updated recommended index parameter set converges, and the recommended index parameter set obtained in the last round of iteration is used as the recommended The target recommendation parameter set output of the information set. In the way of combining local update and aggregate update, iteratively trains the recommendation index parameter set to improve the accuracy of calculating the target recommendation value of recommendation information.
Description
technical field [0001] The invention relates to the field of computers, in particular to a training method and device for an information recommendation system. Background technique [0002] With the development of science and technology, users will receive a large amount of information resources and generate a large amount of data information when using terminals. For example, when users browse e-commerce websites, they will perform feedback behaviors on the products on the websites. The feedback behaviors include Positive behaviors such as bookmarking and sharing products, negative behaviors such as deleting products, and click behaviors of clicking products. How to formulate an efficient and accurate delivery strategy, recommend appropriate information resources to corresponding users, so that users can find content information that may be of interest in a large number of information resources, in order to solve this problem, the existing technology design A centralized r...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/9535G06Q30/06
CPCG06Q30/0631G06F16/9535
Inventor 黄安埠刘洋陈天健杨强
Owner WEBANK (CHINA)