Deep Neural Network Behavioral Biometric Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing behavioral biometric methods rely on manually engineered features, which are fragile and complex, failing to be robust in practice, especially in realistic datasets, and have not fully leveraged the potential of deep learning for improved verification accuracy.
Innovation Solution
An automatic feature extraction framework using deep neural networks to generate determinate vectors, which are used for keystroke and gait-based verification, providing a more robust and efficient method for user identity authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manually engineered features are used for behavioral biometric verification, then domain-specific expertise can be applied, but the features become fragile and overly complex, failing to be robust in practice
Solution Approach 1:
The patent replaces manual feature engineering with automatic feature extraction using deep neural networks. The DNN automatically learns discriminative features from raw behavioral data (keystroke or gait), eliminating the need for handcrafted domain-specific features and reducing complexity while improving robustness through data-driven learning
Solution Approach 2:
The deep neural network performs self-learning by automatically extracting features from raw data without human intervention. The system trains the DNN to autonomously identify discriminative patterns in behavioral data, making the feature extraction process self-sufficient and reducing dependency on manual domain expertise
2Measurement precision
If traditional behavioral biometric methods are used, then existing authentication mechanisms can be enhanced, but they require significant data and fail to achieve low equal error rates in realistic datasets
Solution Approach 1:
The patent changes the fundamental parameters of feature extraction by using deep neural networks with multiple layers of non-linear transformations. This transforms the feature space into a more discriminative representation that achieves lower equal error rates with fewer data samples, fundamentally improving the efficiency of the verification process
3Extent of automation
If deep learning is applied to behavioral biometrics, then automatic feature extraction can be achieved, but the application has not been as pervasive or successful as in other biometric domains
Solution Approach 1:
The patent segments the behavioral verification process into distinct phases: data collection, DNN training, feature extraction, and verification. By dividing the complex task into manageable components and applying targeted deep learning techniques to each stage, the system achieves both high automation and reliable verification accuracy
Solution Approach 2:
The deep neural network transforms raw behavioral data into a higher-dimensional latent feature space where discriminative patterns become more apparent. This dimensional transformation enables the DNN to capture complex temporal and spatial patterns in keystroke and gait data that are not visible in the original feature space, achieving superior verification accuracy
Data Source
AI summary
A method for gait-based behavioral verification of user identity of a subject user of a computer system includes obtaining an enrollment signature corresponding to an identified user and serving as a unique identifier of the identified user, the enrollment signature including an enrollment determinate vector generated based on supplying enrollment gait data to a deep neural network for processing. The method further includes obtaining verification determinate vector(s), the verification determinate vector(s) for comparison to the enrollment signature to determine whether the subject user is the identified user. The method compares the verification determinate vector(s) to the enrollment signature and generates a probability indicator indicating a probability that gait data from a common user produced, from the deep neural network, the enrollment signature and the verification determinate vector(s), and indicates to the computer system whether, based on the probability indicator, the subject user is verified to be the identified user.