Intelligent question answering and process automation method and system across business systems

By constructing a business entity relationship graph and dynamically analyzing user interaction data, the challenges of cross-system intelligent question answering and process automation for AI assistants were solved, enabling dynamic risk assessment and system optimization, and improving the reliability and security of processes.

CN122153017AActive Publication Date: 2026-06-05SHAANXI YULIN ENERGY GRP CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHAANXI YULIN ENERGY GRP CO LTD
Filing Date
2026-05-09
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing AI assistants cannot effectively perform intelligent question answering and reliable process automation across multiple business systems. They lack a deep understanding of the enterprise's overall business logic and cannot dynamically assess the risks of automated operations, resulting in limitations on their level of intelligence and application scope.

Method used

By statically parsing source code to construct a business entity relationship graph, calculating the structural compactness and semantic consistency index, generating time-series curves of user interaction costs and business outcome deviations, identifying cost peak events, quantifying response patterns, calculating performance degradation and problem contribution, generating early warning and handling measures, and realizing dynamic risk perception and optimization.

Benefits of technology

It enables intelligent question answering and process automation across business systems, improves the reliability and security of automated processes, quantifies the root causes of problems and provides precise basis for improvement, and continuously improves system cohesion and availability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122153017A_ABST
    Figure CN122153017A_ABST
Patent Text Reader

Abstract

The application discloses a cross-service system intelligent question answering and process automation method and system, relates to the field of question answering and process automation, and comprises an initial index module, a shock response module, a first adjustment module, a second adjustment module and an early warning processing module. Through calculation of a structure compactness index and a revised semantic consistency index, an initial static logic cohesion index is obtained. A user interaction cost time sequence curve and a business result deviation degree time sequence curve are generated, a cost peak event is identified, and shock response dynamics characteristics are obtained. A total semantic friction coefficient is obtained by weighted summation of a performance degradation index and a first problem contribution degree, a first semantic consistency adjustment value is generated. A total structure friction coefficient is obtained by multiplication and summation of a second problem contribution degree and the performance degradation index, a second structure compactness adjustment value is generated, and early warning processing measures are generated, which can continuously improve the cohesion and availability of the underlying business system.
Need to check novelty before this filing date? Find Prior Art