System and method for evaluation of money transfer patterns

Abstract

A system and method for evaluating records of money transfers for suspicious or irregular transaction patterns. A fraud processing server evaluates money transfer records by blocks, each block consisting of records having the same characteristic relating to location, such as the country where the transfer originates. A second characteristic of the records, such as value, sending agent, or time-of-day, is aggregated and compared to a threshold value. If the threshold value is met, the block of records is analyzed to identify individual records that are suspicious or irregular.

1 . A method for evaluating electronic money transfers, comprising: 
 electronically storing records of money transfer requests, wherein each record is associated with a money transfer request and has at least two data fields, a first data field representing a first characteristic of the money transfer request and a second data field representing a second characteristic of the money transfer request;    sorting the money transfer records to create at least one data block, wherein all records in the data block have the same first characteristic;    calculating a collective value of the second characteristic of all the records in the data block;    comparing the collective value against a predetermined threshold value, threshold value chosen to represent potentially irregular money transfer transactions;    indicating a potentially irregular money transfer pattern if the collective value meets the threshold value; and    analyzing individual records within the block for irregular money transfers if an irregular money transfer pattern has been indicated.    
     
     
         2 . The method of  claim 1 , wherein the first characteristic is related to the location of the money transfer request.  
     
     
         3 . The method of  claim 2 , wherein the first characteristic is the country where the money transfer request in made.  
     
     
         4 . The method of  claim 2 , wherein the first characteristic is the agent network receiving the money transfer request.  
     
     
         5 . The method of  claim 2 , wherein the second characteristic is related to the value of the money transfer request.  
     
     
         6 . The method of  claim 5 , wherein the value of each money transfer request is assigned to one of a plurality of value bands, each band representing a predetermined range of transferred monetary amounts, and wherein the collective value represents the number of money transfer requests having transferred monetary amounts falling within one of the value bands.  
     
     
         7 . The method of  claim 6 , wherein predetermined range has an upper limit below $3000.  
     
     
         8 . The method of  claim 7 , wherein predetermined range has a lower limit above $2800.  
     
     
         9 . The method of  claim 2 , wherein the second characteristic is related to the time-of-day of each transaction.  
     
     
         10 . The method of  claim 9 , wherein the second characteristic is related to whether the transaction is during normal business hours or outside normal business hours, wherein the collective value is the total number of transactions outside normal business hours, and wherein the threshold value is a percentage of the transactions outside normal business hours in relation to the total number of transactions.  
     
     
         11 . The method of  claim 10 , wherein the comparing step compares, for each of a plurality of agents, the collective value of the total number of transactions outside normal business hours to the threshold value.  
     
     
         12 . The method of  claim 2 , wherein the second characteristic is chosen from a group consisting of: 
 transaction value band characteristics;    country corridor characteristics;    agent characteristics;    agent to agent characteristics;    consumer characteristics; and    biographical characteristics.    
     
     
         13 . The method of  claim 1 , wherein the electronic records are stored at a money transfer system, wherein the stored records are transferred to a monitoring system programmed for carrying out the steps of sorting the money transfer records, calculating a collective value for the second characteristic of all the records in the data block, comparing the collective value of the second characteristic against a predetermined threshold value, indicating a potentially irregular money transfer pattern if the collective value of the second characteristic of all records in the data block meets the threshold value, and analyzing individual records within the data bock for irregular money transfers, and wherein the threshold value is a red flag threshold selected by the operator of the money transfer system based on experience.  
     
     
         14 . The method of  claim 1 , wherein the money transfer records include a first sender identification associated with a first money transfer request and at least a second sender identification associated with a second money transfer request, and wherein the step of analyzing individual records comprises: 
 performing an analysis of the records, wherein the analysis indicates the first sender identification and the second sender identification are related;    creating a reference designator, wherein the reference designator is associated with records having the related first and second sender identifications; and    searching the records associated with the reference designator to determine if any of records are suspicious money transfer requests.    
     
     
         15 . The method of  claim 14 , wherein the step of searching includes searching the records to determine if any of the money transfer requests are by a known suspicious user.  
     
     
         16 . A method for evaluating electronic money transfers, comprising: 
 receiving a plurality of money transfer requests;    electronically storing records of the money transfer requests, wherein each record is associated with one money transfer request and has data that defines characteristics of that money transfer request, including at least an location characteristic relating to the location where the money transfer request was made and a transaction characteristic relating to a characteristic other than location;    sorting the money transfer records into at least one block, wherein all records in the block have the same location characteristic;    aggregating the transaction characteristic of all the records in the block to arrive at a collective value for the transaction characteristic;    defining a red flag threshold level for the collective value, the threshold level chosen to represent, if met, a suspicious money transfer pattern;    comparing the collective value for the transactional characteristics with the threshold level;    indicating a suspicious money transfer pattern if the collective value for the transaction characteristic meets the threshold level; and    analyzing individual records within the block if a suspicious money transfer pattern has been indicated.    
     
     
         17 . A system for evaluating money transfer records, comprising: 
 a database for storing a plurality money transfer records, each record having a plurality of data fields relating to the money transfer, including a first field relating to a characteristic of the location where the money transfer was requested and a second field relating to a transaction characteristic not related to location;    a fraud processing server for evaluating the money transfer records, the fraud processing server programmed to: 
 sort the money records into one or more blocks of records, each block having the same characteristic in the first field;  
 aggregate the data in the second field of the records in the block to obtain a collective value of the characteristics in the second field;  
 compare the collective value to a threshold value, wherein the threshold level is chosen to represent, if met, a suspicious money transfer pattern; and  
   indicating a suspicious money transfer pattern if the collective value meets the threshold value, so that the block of records can be further analyzed to identify suspicious individual money transfers.    
     
     
         18 . The system of  claim 17 , wherein the characteristic in the first field identifies the country where the money transfer request in made.  
     
     
         19 . The system of  claim 17 , wherein the characteristic in the first field identifies the agent network processing the money transfer request.  
     
     
         20 . The system of  claim 17 , wherein the characteristic in the second field identifies the value of the money transfer request.  
     
     
         21 . The system of  claim 17 , wherein the characteristic in the second field identifies the time-of-day of each transaction.  
     
     
         22 . The system of  claim 17 , wherein the characteristic in the second field is chosen from a group consisting of: 
 transaction value band characteristics;    country corridor characteristics;    agent characteristics;    agent to agent characteristics;    consumer characteristics; and    biographical characteristics.    
     
     
         23 . The system of  claim 17 , wherein the database is a fraud watch database, wherein records are collected and stored by a host computer and an associated transaction database within an money transfer system in response to money transfer requests by senders of money transfers, and wherein the fraud watch database and fraud processing server are separate from the host computer and transaction database, in order to minimize operational impact on the host computer and transaction database.