System and method detecting fraud using machine-learning and recorded voice clips
AssigneeCapital One Services, LLC
InventorsKaitlin Newman, Jeffrey Rule, Rajko Ilincic
Life of this patent
Estimating the term…
Abstract
A system and method are disclosed for training a machine-learning model to detect characteristics of fraudulent calls. The machine-learning model is trained using audio clips, voice recognition, call handler feedback and general public knowledge of commercial risks to detect and divert fraudulent calls, thereby alleviating the burdens otherwise placed on call center service representatives.
Claims20
A system for detecting fraudulent calls received from an incoming call network by a call service center comprising a plurality of call handlers comprises:
- a processor;
- a fraud training table comprising a fraud training entry storing information related to at least one characteristic of a fraudulent caller as indicated by a call handler, the at least one characteristic including an audible characteristic;
- a machine-learning model coupled to the fraud training table for training the machine-learning model to recognize audible characteristics;
- and
- a call center interface, coupled between the incoming call network and the call handler, the call center interface including program code operable when executed upon by the processor to:
- detect a call comprising an incoming call data stream;
- generate a software agent to monitor the incoming call data stream;
- apply, by the software agent, the machine-learning model to the incoming call data stream to identify audible call characteristics and to determine, based on the audible call characteristics, a fraud probability factor;
- and
- display the fraud probability factor on a display of the call center interface.
A method for detecting fraudulent calls received at a call service center comprises:
- training a machine-learning model to recognize characteristics, including audible characteristics, of fraudulent calls to provide a trained machine-learning model, the training performed using a fraud training table, wherein the fraud training table comprises a plurality of entries corresponding to a plurality of characteristics of fraudulent calls;
- receiving a call comprising an incoming call data stream at the call service center;
- deploying a software call agent to handle the call, the software call agent comprising a copy of the trained machine-learning model;
- during the call, the software call agent analyzing characteristics of the call to identify characteristics of the call that are associated with fraudulent calls utilizing the trained machine-learning model;
- calculating a fraud probability factor that the call is fraudulent in response to the analysis;
- displaying the fraud probability factor on a display of the call service center;
- and
- diverting the call based on the fraud probability factor.
A method for identifying fraudulent calls received at a call service center includes the steps of:
- training a machine-learning model to recognize characteristics, including audible characteristics, of fraudulent calls to provide a trained machine-learning model, the training performed using a fraud training table, wherein the fraud training table comprises a plurality of entries corresponding to a plurality of characteristics of fraudulent calls;
- receiving a call at the call service center;
- deploying a software call agent to handle the call, the software call agent comprising a copy of the trained machine-learning model;
- during the call, calculating a fraud probability factor that the call is fraudulent using the trained machine-learning model;
- displaying the fraud probability factor on a display interface of the call service center, the display interface including an input element enabling a call center agent to modify the fraud probability factor during the call;
- and
- diverting the call based on the fraud probability factor.
Dependencies are inferred from the claim text.