Demand assessment method, device, equipment, storage medium and program product

By combining a large model with preset prompts and a multiple retrieval mechanism, the adaptability and efficiency issues of traditional risk assessment methods are resolved, enabling efficient assessment of new attack techniques and regulatory policies, and improving risk identification and early warning capabilities.

CN122132515APending Publication Date: 2026-06-02INDUSTRIAL AND COMMERCIAL BANK OF CHINA
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
INDUSTRIAL AND COMMERCIAL BANK OF CHINA
Filing Date
2025-08-14
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Traditional risk assessment methods are difficult to adapt to new attack techniques and changes in regulatory policies. They are inefficient and rely on human experience to make it difficult to achieve efficient and comprehensive risk coverage, which can easily lead to business risks.

Method used

By employing a demand assessment method based on a large model, and utilizing pre-defined thought chains indicated by pre-defined prompts, the target large model is guided to conduct progressive reasoning of causal relationships. Combined with multiple searches and content expansion, accurate assessment results are generated.

Benefits of technology

It has improved the breadth and depth of risk identification, reduced assessment bias, enhanced the early warning capability of attack methods, and enabled risk analysis and accurate assessment of complex business links.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122132515A_ABST
    Figure CN122132515A_ABST
Patent Text Reader

Abstract

This application provides a requirement assessment method relating to the fields of artificial intelligence, information security, and the application of large-scale models in fintech. The requirement assessment method includes: responding to a requirement assessment request; retrieving a first matching search result based on a business requirement document to be assessed, the first search result including rules for assessing the requirement content in the business requirement document; retrieving a second matching search result based on at least a portion of the content of the first search result; performing content expansion on at least a portion of the content based on the second search result to obtain first reference data; and guiding a target large-scale model to generate an assessment result of the requirement content based on the first reference data using a preset thought chain indicated by preset prompt words, the preset thought chain including reasoning steps for guiding the target large-scale model to assess the requirement content. This application also provides a requirement assessment apparatus, device, storage medium, and program product.
Need to check novelty before this filing date? Find Prior Art