One-shot Behavioral Biometrics Login Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional password-based authentication methods are vulnerable to attacks, such as brute-forcing and data breaches, and require extensive training data for user authentication, which poses security risks and inefficiencies.
Innovation Solution
A biometrics-based user login authentication system using a Siamese neural network that authenticates users based on behavioral data, such as keyboard strokes and mouse movements, allowing for one-shot authentication without the need for extensive user-specific training data, leveraging domain-specific features for improved accuracy and scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional password-based authentication is used, then implementation is simple and widely compatible, but security is vulnerable to attacks such as brute-forcing and data breaches
Solution Approach 1:
The patent replaces traditional password-based mechanical authentication with behavioral biometric authentication using machine learning models. The system captures behavioral data (keyboard strokes, mouse movements, touch patterns) and uses a Siamese neural network to authenticate users based on behavioral patterns rather than passwords, thereby improving security without requiring complex infrastructure changes
Solution Approach 2:
The patent introduces behavioral biometric data as an intermediary between the user and the authentication system. Instead of directly verifying passwords, the system captures and analyzes behavioral patterns during interaction with the device, using these patterns as a mediator to verify user identity and improve security against traditional attacks
2Measurement precision
If extensive user-specific training data is collected for authentication, then authentication accuracy improves, but security risks and computational resource requirements increase
Solution Approach 1:
The patent performs preliminary action by collecting behavioral data during normal user interaction with the device rather than requiring separate training sessions. The system continuously captures keyboard strokes, mouse movements, and touch patterns during regular usage, preprocessing this data for future authentication without disrupting user workflow or requiring additional data collection efforts
Solution Approach 2:
The system implements self-service by automatically capturing and processing behavioral data during normal device usage. Users do not need to explicitly participate in training or provide additional information - the system autonomously collects behavioral patterns during regular interaction and uses them for authentication, eliminating the need for extensive separate training data collection
3Productivity
If one-shot authentication is implemented, then user convenience and processing speed improve, but authentication reliability may be compromised without extensive training data
Solution Approach 1:
The patent changes the parameters used for authentication from static passwords to dynamic behavioral patterns. By analyzing multiple behavioral dimensions (keyboard stroke dynamics, mouse movement patterns, touch screen interaction characteristics) simultaneously, the system achieves reliable one-shot authentication that captures the uniqueness of each user's interaction style without requiring extensive training data
Solution Approach 2:
The patent adds another dimension to authentication by incorporating temporal and spatial characteristics of user behavior. Instead of relying solely on static credentials, the system analyzes the dynamics of user interactions over time and across different spatial locations on the device, creating a multi-dimensional behavioral signature that enables reliable quick authentication
Data Source
AI summary
In one approach, a method includes: receiving a reference login event input from a user, the reference login event input being associated with a first session of the user logging into an account; receiving a new login event input from the user, the new login event input being associated with a second session of the user logging into the account; accessing a machine learning model, wherein the machine learning model is trained using data selected based on a similarity of behavior between different users; and authenticating, with the machine learning model, the user for the account, based at least in part on the reference login event input and the new login event input. In examples, the reference and new login event inputs comprise one or more items of biometric data generated by interaction of the user in a web environment and/or a mobile environment for logging into the account.