| 摘要 |
When a transaction is made between a transaction device 102 e.g. an ATM or POS terminal the location of a user is determined, for example based on the location of the users mobile phone, and it is determined whether this location can be used to infer the location of the transaction device by comparison with a distribution map or database 104-105. A counter may be increased where the inferred location of the device matches a previously inferred location of the same device, increasing the confidence level that the device is in that location. If the inferred location is different to a previously inferred location it may be that the result is not valid, for example it may be that the user has left their mobile phone in a different location. A threshold confidence level may be used to determine whether location information has been sufficiently reliably established. The distribution map for transaction devices may be automatically generated through a self-learning process. Thus location may be used to assess whether a transaction is fraudulent 101 even when accurate location data is not available for a transaction device. |
