Telephony Fraud Risk Score Using Audio and Non-Audio Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Modern enterprises face challenges in detecting fraudulent activities, particularly credit card fraud and identity-based fraud in job applications, where traditional methods like background checks are ineffective against individuals assuming new identities.

Innovation Solution

A system and method for generating a fraud risk score by analyzing both audio and non-audio channel data, using a telephony risk score calculator and an aggregate risk score generator to determine the likelihood of fraud based on voice characteristics and transaction data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional background checks based on identity factors (names, social security numbers) are used, then the detection process is simple and quick, but they become essentially useless when fraudsters assume new identities

Engineering Contradiction:

Improvefraud detection effectivenessVSAvoiddetection system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces voice biometric analysis as an intermediary detection mechanism that bridges the gap between simple identity verification and complex fraud detection. Instead of directly relying on identity factors that can be falsified, the system uses voice characteristics as a mediator that is harder to replicate, thereby improving reliability without requiring overly complex investigative procedures

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional mechanical/document-based verification methods with acoustic/voice-based verification. By substituting the mechanical process of checking physical documents and identity papers with acoustic analysis of voice patterns, the system achieves more reliable fraud detection that works even when identity documents are forged or identities are assumed

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If audio channel data and non-audio channel data are analyzed together to generate fraud risk scores, then fraud detection accuracy is improved, but the processing complexity and computational requirements increase

Engineering Contradiction:

Improvefraud risk assessment accuracyVSAvoiddata processing system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the fraud detection process into distinct analytical components: audio channel analysis (voice biometrics, tone, pitch) and non-audio channel analysis (transaction patterns, call metadata). Each segment processes specific types of data independently, then their results are integrated to form the comprehensive fraud risk score. This segmentation allows complex multi-source data to be processed in manageable, specialized modules rather than requiring a monolithic complex system

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent creates a multi-functional risk scoring system that can process and analyze multiple types of data channels (audio and non-audio) through a unified framework. The same risk scoring methodology is applied across different data types, allowing the system to handle diverse inputs through a single versatile processing architecture, thereby managing complexity through standardization rather than requiring separate specialized systems for each data type

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If voice biometric analysis is used to identify fraudsters, then the ability to detect identity-based fraud is improved, but the time required for analysis increases

Engineering Contradiction:

Improveidentity fraud detection capabilityVSAvoidanalysis time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts and analyzes voice biometric features during the initial phone conversation itself, rather than requiring a separate analysis phase. Voiceprints, pitch patterns, and other acoustic characteristics are captured and processed in real-time as part of the normal interaction, allowing fraud detection to occur concurrently with communication rather than adding sequential time delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system performs partial voice biometric analysis during the call by focusing on key acoustic features (pitch, tone, voiceprint) rather than attempting to analyze every aspect of the voice signal. This selective extraction of the most discriminative features enables sufficient fraud detection capability with reduced processing time compared to comprehensive voice analysis

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS8930261B2Method and system for generating a fraud risk score using telephony channel based audio and non-audio data

Publication Date: 2015.01.06 VERINT AMERICAS INC

AI summary

Disclosed is a method for generating a fraud risk score representing a fraud risk associated with an individual, the method comprising: a) determining a telephony channel risk score from at least one of audio channel data and non-audio channel data of the individual; and b) generating the fraud risk score based on at least one of the telephony channel risk score, the audio channel data, and the non-audio channel data.