Fraud detection database

• US 9,589,566 B2
• Filed: 03/21/2014
• Issued: 03/07/2017
• Est. Priority Date: 03/21/2014
• Status: Active Grant

• Current Assignee: Wells Fargo Bank NA (Wells Fargo & Company)
• Original Assignee: Wells Fargo Bank NA (Wells Fargo & Company)
• Inventors: Jones, Raymond F.
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US14/222,188
• Publication Number:
  US 20150269946A1
• Time in Patent Office: 1,082 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06Q 30/0185 Product, service or business identity fraud
• CPC Class Codes: G10L 17/02 Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 17/26 Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: H04M 15/47 Fraud detection or prevention means
• CPC Class Codes: H04M 2203/301 Management of recordings
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

Embodiments of techniques or systems for fraud detection are provided herein. A communication may be received where the communication includes one or more voice signals from an individual. Frequency responses associated with these voice signals may be determined and analyzed and utilized to determine whether or not potential fraudulent activity is occurring. For example, if a frequency response is greater than a frequency threshold, potential fraudulent activity may be determined. Further, frequency responses may be cross referenced with voice biometrics, voice printing, or fraud pathway detection results. In this way, voice stress or frequency responses may be utilized to build other databases related to other types of fraud detection, thereby enhancing one or more aspects of fraud detection. For example, a database may include a voice library, a pathway library, or a frequency library which include characteristics associated with fraudulent activity, thereby facilitating identification of such activity.

Claims

1. A system for fraud detection, comprising:

a processing unit that executes the following computer executable components stored in a memory;

a database component comprising;

a voice library comprising one or more voice samples of individuals deemed to be fraudsters, and

a pathway library comprising a set of characteristics previously identified as being associated with a fraudulent communication, the set of characteristics are associated with artifacts detected during portions of the fraudulent communication that lack voice signals;

a monitoring component that receives a communication that comprises voice signals of an individual making requests related to a financial account;

a detection component that determines characteristics associated with the communication, the characteristics relate to at least one artifact of the communication detected during a portion without a voice signal;

an analysis component that determines if a difference between a first frequency of a first segment of the communication and a second frequency of a second segment of the communication is outside a frequency range provided by a frequency library, wherein the first frequency is weighted more heavily than the second frequency based on a first type of communication segment assigned to the first segment and a second type of communication segment assigned to the second segment; and

a fraud component that determines activity associated with the financial account is fraudulent based on a determination that the at least one detected artifact of the communication matches the artifacts detected during portions of the fraudulent communication and that the difference between the first frequency and the second frequency is outside of the frequency range.

2. The system of claim 1, wherein the analysis component determines respective frequency responses of the voice signals of the communication.

3. The system of claim 2, wherein the fraud component updates the frequency library with one or more of the frequency responses associated with the individual based on the determination that the communication is the fraudulent communication.

4. The system of claim 1, wherein the fraud component deems the activity associated with the financial account is fraudulent based on a comparison between the voice signals and at least one voice sample of the one or more of the voice samples of the voice library.

5. The system of claim 1, wherein the activity deemed to be fraudulent for the financial account is determined at substantially the same time as the communication is in process.

6. A method for fraud detection, comprising:

receiving, by a system comprising a processing unit, a communication that comprises voice signals of an individual making one or more requests related to an account;

determining, by the system, an artifact of the communication, the artifact being detected during a portion of the communication void of voice signals;

analyzing, by the system, a frequency response of the voice signals of the individual;

applying, by the system, a first weight to a first frequency response of a first segment of the voice signals based on an identified communication type of the first segment, and a second weight to a second frequency response of a second segment of the voice signals based on another identified communication type of the second segment, wherein the first weight and the second weight are different weights;

determining, by the system, a frequency difference between a frequency response of the first segment of the voice signals and the second segment of the voice signals;

determining, by the system, the frequency difference exceeds a frequency threshold;

determining, by the system, the individual is engaging in fraudulent activity based on a determination that the frequency difference exceeds the frequency threshold and another determination that the artifact matches at least one artifact previously identified as being associated with a fraudulent communication and retained in a pathway library; and

updating, by the system, the pathway library to include the artifact of the communication based on the determination that the individual is engaging in fraudulent activity.

7. The method of claim 6, further comprises updating, by the system, a frequency library to include the frequency response associated with the individual based on the determination that the individual is engaging in fraudulent activity.

8. The method of claim 6, further comprises determining, by the system, activity associated with the account is fraudulent based on a match between the voice signals and one or more voice samples of a voice library.

9. The method of claim 6, further comprises determining, by the system, activity associated with the account is fraudulent based on a match between one or more characteristics associated with the communication and one or more sets of characteristics included in the pathway library.

10. A system for fraud detection, comprising:

a processing unit that executes the following computer executable components stored in a memory;

a database component comprising;

a voice library comprising one or more voice samples of individuals deemed to be fraudsters; and

a pathway library comprising one or more sets of characteristics deemed to be associated with fraudulent communication;

a monitoring component that receives a communication that comprises voice signals of an individual making one or more requests associated with an account;

a detection component that determines a set of characteristics associated with the communication, a characteristic of the set of characteristics is an artifact identified in the absence of the voice signals;

an analysis component that analyzes a frequency response of a plurality of test questions, wherein each of the plurality of test questions is associated with a different frequency threshold and is associated with a different communication segment type; and

a fraud component that indicates the communication is fraudulent based on a determination that at least one frequency response of the plurality of test questions does not match an expected frequency response for the respective communication segment type and that the artifact matches at least one artifact previously identified as being associated with a fraudulent communication and retained in the pathway library, the fraud component updates the voice library or the pathway library based on activity deemed to be fraudulent associated with the account.

11. The system of claim 10, wherein the fraud component updates the voice library to include the voice signals based on a match between the one or more characteristics associated with the communication and the one or more sets characteristics of the pathway library.

12. The system of claim 10, wherein the fraud component updates the voice library to include the voice signals based on one or more frequency responses of the one or more voice signals of the communication.

13. The system of claim 10, wherein the fraud component updates the pathway library to include the one or more characteristics of the communication based on a match between the one or more voice signals and the one or more voice samples of the voice library.

14. The system of claim 10, wherein the fraud component updates the pathway library to include the one or more characteristics of the communication based on one or more frequency responses of the one or more voice signals of the communication.

15. The system of claim 10, wherein the analysis component determines one or more frequency responses of the one or more voice signals of the communication.

16. The system of claim 10, wherein the fraud component updates a frequency library to include the frequency response associated with the individual.

17. The system of claim 10, the monitoring component filters noise from one or more segments of the communication.