Method and system for intelligent detection of service anomaly based on time series prediction

CN122241188APending Publication Date: 2026-06-19GUANGZHOU LESHUI INFORMATION TECH CO LTD

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: GUANGZHOU LESHUI INFORMATION TECH CO LTD
Filing Date: 2026-03-30
Publication Date: 2026-06-19

AI Technical Summary

Technical Problem

Existing technologies cannot effectively integrate continuous time-series business data from multiple modules in an enterprise management system, nor can they construct a model that comprehensively reflects the collaborative evolution of enterprise business. This results in an inability to accurately predict the collaborative evolution trend of multi-module business and to promptly detect business problems caused by abnormal collaboration between modules.

Method used

Collect continuous time-series business data from multiple modules covered by the enterprise management system, construct a cross-module business time-series collaborative evolution network, integrate the business evolution patterns of each module and cross-module collaborative logic based on the AI collaborative evolution prediction model, collect current business operation data in real time and trace deviations, and generate business anomaly detection results.

Benefits of technology

By accurately identifying nodes associated with enterprise management anomalies, the ability of enterprises to respond to business anomalies is improved, ensuring the stability and efficiency of enterprise operation services.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122241188A_ABST

Patent Text Reader

Abstract

This invention provides a business anomaly intelligent detection method and system based on time-series prediction. First, it collects continuous time-series business data from multiple modules covered by an enterprise management system, including data from human resource management, financial management, operations management, and collaborative office work. This data is then mapped to the enterprise management scenario dimension to construct a cross-module business time-series collaborative evolution network. Next, based on this network, an AI collaborative evolution prediction model is built, outputting multi-module collaborative evolution trend information corresponding to the current business operation. Then, current business operation data is collected in real time and adapted to the model to trace the entire chain of deviations, locking down anomaly-related nodes and cross-module collaborative anomaly characteristics. Finally, a business anomaly detection result is generated, including anomaly node location, collaborative deviation description, related impact range, and intelligent handling suggestions, which is synchronized to the decision-making module of the enterprise management system. This invention can comprehensively and accurately detect business anomalies.

Need to check novelty before this filing date? Find Prior Art