Industrial chain enterprise intelligent matching method and system

By using a large language model and dynamic weighted feature similarity calculation, the problem of lag in determining the relationship between the industrial chain and enterprises is solved, achieving real-time and accurate intelligent matching and improving accuracy.

CN121365091APending Publication Date: 2026-01-20IRESEARCH CLOUD TECHNOLOGY (HANGZHOU) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511532416.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-10-24
Publication Date
2026-01-20

AI Technical Summary

Technical Problem

Existing technologies lag behind in determining relationships between supply chains and enterprises, failing to process the impact of dynamic information in real time, leading to inaccurate relationship calculations.

Method used

By employing large language models and semantic recognition technology, combined with dynamic weighted feature similarity calculation formulas, and utilizing real-time information and multi-dimensional features, the correlation between the industrial chain and enterprises is calculated, including time decay factors, market trends, and expert experience parameters, to achieve intelligent matching.

Benefits of technology

It improves the accuracy and real-time performance of determining the relationship between the industrial chain and enterprises, with an average accuracy rate of 95.83%, which is better than the 76.3% of existing technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121365091A_ABST
    Figure CN121365091A_ABST
Patent Text Reader

Abstract

The invention discloses an industry chain enterprise intelligent matching method, and belongs to the technical field of computers. The method comprises the following steps: acquiring enterprise and product information; the enterprise and product information comprises an enterprise name and a product name; inputting enterprise and product information into an execution planning module, and outputting an execution plan; inputting the execution plan into a knowledge base search module for searching to obtain industrial chain feature information; inputting the execution plan into a public domain information search module for searching to obtain enterprise feature information; and calculating the association degree between the industry chain feature information and the enterprise feature information. According to the industrial chain and enterprise relationship mining method provided by the invention, a dynamic weight feature similarity calculation formula is introduced, and intelligent weighting is realized in combination with multi-dimensional features, market trends and time decay factors of the industrial chain and the enterprise.
Need to check novelty before this filing date? Find Prior Art