An event context generation method, apparatus, device, medium and program product
By dividing the popularity intervals based on the relevance of candidate texts and their publication time, and selecting target texts to generate event timelines, the problem of sampling strategies relying on experience in existing technologies is solved, thereby improving the accuracy and stability of event timelines.
CN122309734APending Publication Date: 2026-06-30BEIJING ZITIAO NETWORK TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING ZITIAO NETWORK TECH CO LTD
- Filing Date
- 2024-12-30
- Publication Date
- 2026-06-30
AI Technical Summary
Technical Problem
Existing technologies require extensive experience in developing sampling strategies for large amounts of text when generating event timelines, and inappropriate sampling strategies can affect the accuracy of event timelines.
Method used
By acquiring multiple candidate texts, determining the candidate text set based on the degree of correlation between the candidate texts and the publication time, dividing the popularity interval based on text attributes and the preset number of target texts, selecting target texts that meet the preset set conditions for sampling, and finally generating the event timeline.
Benefits of technology
It improves the accuracy and stability of event timelines, reduces computational resource consumption, and shortens generation time.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122309734A_ABST
Abstract
This disclosure provides an event context generation method, apparatus, device, medium, and program product. The method includes: acquiring multiple candidate texts; determining at least one candidate text set based on the correlation between the candidate texts; determining multiple popularity intervals based on the publication time of the candidate texts in each candidate text set, a preset number of target texts, and the text attributes of the candidate texts; for each popularity interval, determining target texts within the popularity interval based on the target text set corresponding to the candidate texts within the popularity interval; and generating an event context based on the event association information of each target text. The technical solution of this disclosure, by determining target texts based on the target text set corresponding to the candidate texts within the popularity interval, achieves stable sampling of candidate texts in each popularity interval to obtain target texts based on popularity, thereby improving the stability and accuracy of the event context.
Need to check novelty before this filing date? Find Prior Art