Media data acquisition method and system for constructing spatio-temporal knowledge tree based on hint engineering

By constructing a spatiotemporal knowledge tree based on prompting engineering and utilizing a large language model and a controllable prompting strategy, the problems of spatiotemporal correlation and low annotation cost in social media data acquisition are solved, achieving efficient and accurate data acquisition and event tracking, and is suitable for intelligent analysis in various scenarios.

CN121960464BActive Publication Date: 2026-06-26MOGANSHAN DIXIN LABORATORY

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
MOGANSHAN DIXIN LABORATORY
Filing Date
2026-04-02
Publication Date
2026-06-26

Smart Images

  • Figure CN121960464B_ABST
    Figure CN121960464B_ABST
Patent Text Reader

Abstract

The application belongs to the field of artificial intelligence and information technology, and specifically relates to a media data acquisition method and system for constructing a space-time knowledge tree based on a hinting project. The system comprises a hinting project module, a common sense knowledge tree construction module, a space-time knowledge tree construction module and a media data acquisition module. The hinting project module provides a hint word template. The common sense knowledge tree construction module constructs a common sense knowledge tree through large language model reasoning. The space-time knowledge tree construction module calls a place refinement hint word template and a facility refinement hint word template based on the common sense knowledge tree, and constructs a space-time knowledge tree through large language model reasoning and multi-element geographic information verification. The media data acquisition module generates multi-dimensional search conditions to drive a data acquisition engine to acquire data. The method uses large model reasoning and controllable hinting strategies to realize automatic construction and dynamic expansion of the knowledge tree, and drives the crawler to perform multi-dimensional accurate search, so that high-value data can be accurately captured and evolution clues can be reliably tracked.
Need to check novelty before this filing date? Find Prior Art