Anomaly analysis method, device, medium, and product for financial data

By constructing a financial knowledge graph and combining it with anomaly detection algorithms, the system automatically identifies financial data anomalies and traces their root causes in business operations. This solves the problems of insufficient analytical depth and rigid report generation in existing technologies, and achieves accurate identification and in-depth analysis of financial data.

CN122241062APending Publication Date: 2026-06-19KINGDEE SOFTWARE(CHINA) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
KINGDEE SOFTWARE(CHINA) CO LTD
Filing Date
2026-02-28
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing financial analysis systems are unable to delve into the business-level factors that cause anomalies in financial data, resulting in analysis results that remain superficial and fail to meet the company's need to explore deeper causes.

Method used

By constructing a financial knowledge graph and combining a hybrid anomaly detection mechanism of rules and anomaly detection algorithms, the system can achieve comprehensive and accurate identification of financial data and automatically generate analysis reports. The system can also automatically perform penetrating attribution analysis along the pre-set levels and relationships in the knowledge graph to locate the root cause node in the specific business dimension.

Benefits of technology

It significantly improves the efficiency of financial analysis and the accuracy of anomaly analysis, solving the problems of single anomaly detection methods, insufficient analysis depth, and rigid report generation in traditional solutions, and realizing comprehensive identification and in-depth tracing of financial data anomalies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122241062A_ABST
    Figure CN122241062A_ABST
Patent Text Reader

Abstract

This application relates to the field of computer technology, and in particular to a method, device, medium, and product for anomaly analysis of financial data. The method includes: acquiring a financial knowledge graph of the object to be analyzed, the financial knowledge graph including hierarchical relationships between multiple financial subjects and multiple business dimensions associated with each financial subject; acquiring financial data of the object to be analyzed within a target time period; performing anomaly detection on the financial data within the target time period, identifying anomalous financial subjects from among the multiple financial subjects; determining the associated business dimensions of the anomalous financial subjects based on the multiple business dimensions associated with each financial subject; performing attribution analysis based on the associated business dimensions and the financial knowledge graph to obtain attribution chains; and generating an analysis report based on the attribution chains. This method can accurately identify and locate anomalies in financial data, meeting the need for in-depth investigation of the causes of financial data anomalies.
Need to check novelty before this filing date? Find Prior Art