System architecture for fraud detection

A method of detecting fraud in a client contact system, comprising:

• receiving a contact from a user, the contact corresponding to event data, the event data including at least one of a channel address and an identifier;
• processing the channel address to determine if the channel address is whitelisted or not whitelisted, and, upon determining that the channel address is whitelisted, allowing the contact to proceed without further processing if the channel address is whitelisted;
• upon determining that the channel address is not whitelisted, aggregating records to score the event data by checking the event data against criteria to determine a risk associated with the event data, wherein the criteria comprises at least one of behavior criterion, situation criterion and reputation criterion, where the criteria are based on the aggregated records;
• and wherein the behavior criterion is determined by calculating a statistical standard deviation over a predetermined period of time for a typical contact with the client contact system and determining whether the event data falls outside the statistical standard deviation;
• outputting a score based on the criteria and passing the contact from the user to the client contact system when the score is within a predetermined range;
• and
• updating the aggregated records according to the criteria for the channel address and identifier.