Micro-expression Analysis for User Authenticity Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining authenticity in electronic transactions are limited by relying on transaction or credit history, which can be fabricated, leading to inaccurate assessments and increased costs due to the need for face-to-face interactions or phone conversations.
Innovation Solution
A system using convolutional neural networks to analyze video and audio data for micro-expressions, generating questions to collect additional data and determining authenticity based on predicted emotions, allowing for accurate and efficient electronic assessment of user authenticity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods use transaction or credit history to evaluate credibility, then the assessment process is simple, but the accuracy is low because the data can be fabricated
Solution Approach 1:
The patent replaces traditional mechanical verification methods (checking transaction history, credit reports, and documents) with a biologically-inspired neural network system that processes video and audio data. The CNN-based emotion recognition system analyzes facial expressions, voice tone, and micro-expressions to determine authenticity, substituting manual document verification with automated biometric and behavioral analysis.
Solution Approach 2:
The patent introduces an intermediary system consisting of the convolutional neural network and emotion recognition algorithms that mediate between the user's video/audio input and the authenticity determination. This intermediary processes the raw data through multiple layers of analysis, extracting meaningful patterns and emotions that directly indicate authenticity, thereby improving accuracy while maintaining system manageability.
2Measurement precision
If face-to-face interaction or phone conversation is used to verify authenticity, then the accuracy improves, but the transaction cost increases and the process is delayed
Solution Approach 1:
The system enables self-service authenticity verification where users independently complete the verification process by recording video and audio responses to questions. The neural network automatically analyzes the data and determines authenticity without requiring human operators to conduct interviews or verify documents, thereby maintaining high accuracy while dramatically improving processing efficiency and reducing costs.
Solution Approach 2:
The patent performs preliminary action by having users record their video and audio responses in advance before the actual transaction occurs. The system processes this pre-collected data through the neural network to determine authenticity upfront, eliminating the need for time-consuming face-to-face interactions during the transaction process and thereby improving overall productivity.
3Measurement precision
If face-to-face interaction is required to determine authenticity, then the accuracy improves, but the transaction cost increases
Solution Approach 1:
The patent replaces expensive human-operated face-to-face verification with an automated neural network system that processes video and audio data. This substitution maintains high accuracy in authenticity determination while eliminating the costs associated with human time, travel, and facility requirements, thereby significantly reducing transaction costs.
Solution Approach 2:
The system creates a digital copy of the face-to-face interaction experience by analyzing video and audio recordings through the neural network. This digital copy captures the same authenticity indicators (facial expressions, voice tone, micro-expressions) that would be observed in person, maintaining verification accuracy while eliminating the need for physical presence and associated costs.
Data Source
AI summary
Systems and methods are provided for calculating authenticity of a human user. One method comprises receiving, via a network, an electronic request from a user device, instantiating a video connection with the user device; generating, using a database of questions, a first question; providing, via the network, the generated question to the user device; analyzing video and audio data received via the connection to extract facial expressions, calculating, using convolutional neural networks, first data and second data corresponding predetermined emotions based on facial expressions and audio data; generating candidate emotion data using the first and second data; determining whether the candidate emotion data predicts a predetermined emotion, and generating a second question to collect additional data for aggregating with the first and second data or determining the authenticity of the user and using the determined authenticity to decide on the user request.