POS machine transaction risk identification method based on data analysis and deep learning technology

By using data analysis and deep learning methods, POS transaction data is collected in real time and multi-dimensional features are extracted. A hybrid deep learning model is used for risk scoring and judgment, which solves the problem of insufficient accuracy and timeliness of risk identification in existing technologies and achieves more efficient risk identification and handling.

CN120471624BActive Publication Date: 2026-06-05深圳市阿龙电子有限公司
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
深圳市阿龙电子有限公司
Filing Date
2025-04-29
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing POS transaction risk identification methods rely on experience and fixed rules, which cannot adapt to complex and ever-changing risk scenarios. Deep learning models do not delve deeply enough into data feature mining and have insufficient generalization ability, resulting in a large number of missed and false judgments.

Method used

Using a data analysis and deep learning approach, POS machine transaction data is collected in real time, multi-dimensional feature extraction and dynamic temporal segmentation are performed, and risk scoring and judgment are performed using a hybrid model of temporal convolutional networks with multi-head attention mechanism and gated recurrent units and graph neural networks. Fine-grained feature enhancement is combined with geographical location, device fingerprint and user historical transaction patterns to trigger real-time risk handling instructions.

Benefits of technology

It improves the accuracy, timeliness, and comprehensiveness of transaction risk identification, enabling the timely detection of potential risks and the implementation of effective measures to reduce risk losses.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120471624B_ABST
    Figure CN120471624B_ABST
Patent Text Reader

Abstract

The application relates to the technical field of financial transaction security, and discloses a POS machine transaction risk identification method based on data analysis and deep learning technology. The method collects transaction data streams in real time, inputs a first deep learning model for preliminary risk scoring after multidimensional feature extraction and dynamic time sequence segmentation, screens high-risk time windows, enhances the transaction data fine-grained features of the high-risk time windows, inputs a second deep learning model for secondary risk determination, generates a transaction risk label, and triggers a real-time risk disposal instruction based on the transaction risk label. The multidimensional feature extraction comprehensively depicts the transaction data, the deep learning model effectively mines the risk features, and the dynamic time sequence segmentation is suitable for different transaction frequencies. The application can accurately identify the POS machine transaction risk, timely take disposal measures, guarantee the safety of financial transactions, and has remarkable effects in improving the risk identification accuracy, timeliness and comprehensiveness.
Need to check novelty before this filing date? Find Prior Art