Abnormal Transaction Detection System

Abstract

An abnormal transaction detection system includes a processing unit, a communication unit, and a storage unit electrically connected to the processing unit. When the processing unit receives a transaction request from a user terminal device via the communication unit, including transaction data and physiological monitoring data, the processing unit processes the transaction data and the physiological monitoring data through a data preprocessing procedure and then inputs them into an abnormal transaction detection model to generate a risk identification result. When the processing unit determines that the risk identification result is a high-risk result, it transmits a transaction anomaly notification to the user terminal device via the communication unit.

Claims

1. An abnormal transaction detection system, applicable to a user terminal device, comprising: a communication unit for providing networking functionality; a storage unit for storing a data preprocessing program and an abnormal transaction detection model completed through machine learning; and a processing unit electrically connected to the communication unit and the storage unit, and communicating with the user terminal device through the communication unit; wherein, When the processing unit receives a transaction request from the user terminal device via the communication unit, which includes transaction data and physiological monitoring data, the transaction data and physiological monitoring data are respectively processed by the data preprocessing program and then input into the abnormal transaction detection model to generate a risk identification result; when the processing unit determines that the risk identification result is a high-risk identification result, it transmits a transaction anomaly notification to the user terminal device via the communication unit.

2. The abnormal transaction detection system as described in claim 1, wherein, In this preprocessing procedure, the processing unit uses a statistical method to identify outliers to identify and delete outliers in the transaction data and the physiological monitoring data, and then standardizes the transaction data and the physiological monitoring data after deleting outliers.

3. The abnormal transaction detection system as described in claim 1, wherein, The storage unit also stores a transaction interface. When the processing unit provides the transaction interface to the user terminal device and determines that the transaction request received from the user terminal device does not include the physiological monitoring data, the processing unit locks the transaction request and sends an alarm notification to the user terminal device through the communication unit.

4. The abnormal transaction detection system as described in claim 1, wherein, The storage unit also stores a preset schedule and historical physiological monitoring data of multiple low-risk transactions within a fixed time interval. When the processing unit determines that the risk identification result is a low-risk identification result, it stores the physiological monitoring data in the storage unit. According to the preset schedule, the physiological monitoring data and the historical physiological monitoring data are respectively processed by the data preprocessing procedure and then input into the abnormal transaction detection model to perform and complete machine learning training.

5. The abnormal transaction detection system as described in claim 1, wherein, The storage unit also stores multiple historical physiological monitoring data of low-risk transactions within a fixed time interval. When the processing unit receives a calibration request from the user terminal device that includes calibration physiological data, it performs the data preprocessing procedure on the calibration physiological data and the historical physiological monitoring data respectively, and then inputs them into the abnormal transaction detection model to perform and complete machine learning training.

6. The abnormal transaction detection system as described in claim 1, wherein, The transaction data includes at least one of the following: transaction amount, transaction frequency, transaction time, transaction location, operational behavior data, and social behavior data. The operational behavior data includes operation speed information, number of input errors, and transaction type. The social behavior data includes communication information.

7. The abnormal transaction detection system as described in claim 1, wherein, The physiological monitoring data includes at least one of the following: heart rate information, skin conductance information, blood pressure information, and gyroscope information.

Images