A data processing method, apparatus and device

By constructing a tree-like knowledge graph to expand relevant nodes both vertically and horizontally, the problem of incomplete candidate object screening in existing technologies is solved, resulting in more accurate and comprehensive search or recommendation results and improving user experience.

CN115795136BActive Publication Date: 2026-07-21RAJAX NETWORK &TECHNOLOGY (SHANGHAI) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
RAJAX NETWORK &TECHNOLOGY (SHANGHAI) CO LTD
Filing Date
2022-12-26
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing technologies, when using knowledge graphs for the recall phase, do not comprehensively screen candidates and cannot cover more objects related to keywords, resulting in inaccurate and incomplete search or recommendation results.

Method used

Construct a tree-structured knowledge graph for the target business scenario. By expanding the knowledge graph vertically and horizontally, identify more relevant nodes. Combine the candidate node set with the candidate objects from the database to filter candidate objects. Utilize the relevance conditions of the vertical and horizontal nodes to filter out more objects related to the keywords.

Benefits of technology

It improves the accuracy and comprehensiveness of search or recommendation results, provides more choices, and enhances the user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115795136B_ABST
    Figure CN115795136B_ABST
Patent Text Reader

Abstract

The embodiment of the specification provides a data processing method. In a recommendation or retrieval scene, a tree-shaped knowledge graph of a target business scene can be constructed in advance. After a target node matched with a keyword is determined in the knowledge graph, more nodes associated with the target node can be further determined based on a hierarchical structure of the tree-shaped knowledge graph, so that a candidate node set related to the keyword can be obtained, and then candidate objects related to the keyword can be screened from a database by using the candidate node set. In this way, more objects related to the keyword can be screened out, so that more choices can be provided for a user, and the retrieval result or the recommendation result is more accurate and more comprehensive.
Need to check novelty before this filing date? Find Prior Art