Object recommendation method and device
By using recall words in the artificial intelligence recommendation system to directly recommend target objects, the problem of the existing technology that cannot update and recommend new objects in real time is solved, and efficient and accurate object recommendation is achieved.
CN114780819BActive Publication Date: 2025-09-30BEIJING BAIDU NETCOM SCI & TECH CO LTD
2 Cites 0 Cited by
Patent Information
- Application Number
- CN202210194594.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-01
- Publication Date
- 2025-09-30
- Estimated Expiration
- 2042-03-01
AI Technical Summary
Technical Problem
Existing AI-based recommendation technologies are difficult to update and recommend newly emerging target objects in real time, and training recommendation models requires a lot of manual labeling and time, resulting in the inability to recommend objects that match users in a timely manner.
Method used
By obtaining the recall words of the target object and adding them to the whitelist, the whitelist is used to filter user requests that match the recall words and directly recommend them to the corresponding users, avoiding the step of training the recommendation model.
Benefits of technology
It achieves real-time and accurate recommendations for newly appeared target objects, reduces the amount of data processing, and improves the efficiency and accuracy of recommendations.
✦ Generated by Eureka AI based on patent content.
Abstract
The present disclosure provides an object recommendation method and apparatus, relating to the field of computer technology, and in particular, to the field of recommendation technology based on artificial intelligence. The implementation scheme comprises: in response to obtaining a target object to be recommended, obtaining a recall word for the target object, and adding the recall word for the target object to a whitelist; in response to obtaining multiple current requests from multiple users, using the whitelist, filtering the multiple current requests to obtain a target current request set, each target current request in the target current request set matching a recall word for the target object in the multiple current requests; and recommending the target object to a target user set among the multiple users corresponding to the target current request set.
Need to check novelty before this filing date? Find Prior Art