Transaction data processing method and apparatus

By employing a hierarchical risk identification approach, combined with risk rules, risk identification models, and large language models, preliminary and in-depth analysis of transaction data is conducted. This addresses the issues of rigid rules and poor model interpretability in existing technologies, achieving more efficient and accurate risk identification.

CN122155722APending Publication Date: 2026-06-05BEIJING PACTERA JINXIN TECH LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING PACTERA JINXIN TECH LTD
Filing Date
2026-01-20
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

In the field of finance, existing technologies for identifying transaction risks during transactions suffer from limitations. Expert rule-based methods are rigid and struggle to cope with rapidly changing fraud patterns, while machine learning models are unable to utilize unstructured textual information and complex fraud patterns, resulting in limited identification capabilities and poor interpretability.

Method used

A hierarchical risk identification approach is adopted. First, a preliminary screening is conducted by setting risk identification rules. Then, a risk identification model is used for secondary identification. For transaction data with uncertain risk levels, a large language model is used for in-depth analysis, incorporating contextual information from the transaction data to improve the accuracy and comprehensiveness of identification.

Benefits of technology

It improves the accuracy and comprehensiveness of risk assessment, reduces the consumption of computing resources, avoids the waste of computing power in large language models, improves processing efficiency, and enhances the accuracy and interpretability of risk identification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122155722A_ABST
    Figure CN122155722A_ABST
Patent Text Reader

Abstract

The present disclosure provides a transaction data processing method and device, the method comprising: receiving a transaction request sent by a user account; preprocessing original transaction data in the transaction request to obtain target transaction data; wherein the target transaction data comprises context data of the original transaction data; performing preliminary risk identification on the target transaction data according to a set risk identification rule to obtain a preliminary risk identification result; in response to the preliminary risk identification result indicating that the target transaction data is normal transaction data, performing secondary risk identification on the target transaction data using a risk identification model to obtain a secondary risk identification result; and in response to the secondary risk identification result indicating that the risk level of the target transaction data is uncertain, performing risk analysis on the target transaction data using a large language model to generate a risk analysis result of the target transaction data; through hierarchical and progressive risk identification and analysis of transaction data, the accuracy and comprehensiveness of risk judgment can be improved.
Need to check novelty before this filing date? Find Prior Art