Free Text Behavioral Biometric Authentication Using ITAD Metric
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing keystroke dynamics systems face challenges in authenticating users in uncontrolled free-text environments, requiring large numbers of keystrokes and struggling with varying user behavior, which hampers their usability and effectiveness in continuous authentication.
Innovation Solution
The implementation of an instance-based tail area density (ITAD) metric and feature-level fusion of keystroke dynamics, using monographs and digraphs to create a cumulative density function-based similarity score for rapid user authentication, allowing for minimal keystroke input and improved security in account recovery mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If distribution-based algorithms are used for keystroke dynamics authentication, then authentication accuracy can be improved with large samples, but the number of keystrokes required increases to 500 or more
Solution Approach 1:
The patent replaces distribution-based statistical algorithms with an instance-based information retrieval approach. Instead of computing complex distribution comparisons across large datasets, the system uses simple instance matching between test keystrokes and profile keystrokes, achieving accurate authentication with minimal keystrokes (as few as 10-20).
Solution Approach 2:
The patent changes the fundamental parameter from requiring hundreds of keystrokes to requiring only a handful. This is achieved by transforming the authentication approach from distribution-based statistical analysis to instance-based similarity comparison, fundamentally altering the sample size requirement parameter.
2Measurement precision
If fixed-text keystroke dynamics are used, then strong authentication performance can be achieved, but the system lacks adaptability to uncontrolled free-text environments
Solution Approach 1:
The patent creates a universal authentication system that works across multiple text environments (fixed-text, controlled free-text, and uncontrolled free-text). The instance-based approach is equally effective whether authenticating passwords, form fields, or natural typing, making the system multi-functional and highly adaptable.
Solution Approach 2:
Instead of constraining users to fixed-text inputs for accurate authentication, the patent inverts the approach by allowing complete freedom in text input while maintaining strong authentication performance. The system adapts to user behavior rather than forcing users to conform to rigid input constraints.
3Reliability
If large numbers of keystrokes are collected for authentication, then better user profiles can be built, but authentication speed decreases and usability is hampered
Solution Approach 1:
The patent applies partial action by using only a small subset of keystrokes (10-20 instances) rather than requiring complete profiles built from hundreds of keystrokes. The instance-based method achieves reliable authentication with partial data, eliminating the need to wait for extensive typing before authentication can occur.
4Reliability
If continuous authentication is implemented in uncontrolled free-text settings, then security can be enhanced, but varying user behavior across activities reduces measurement precision
Solution Approach 1:
The patent implements a dynamic authentication system that adapts to changing user behaviors across different activities. The instance-based approach naturally handles behavioral variations by comparing individual keystroke instances rather than relying on consistent typing patterns, enabling reliable continuous authentication despite changes in typing context, content, or activity.
Data Source
AI summary
A system and method to authenticate users on a computing system using a free text behavioral biometric method by recording on the computer system a dataset for each user to be authenticated to create a user profile for each user to be authenticated, each data set comprising a plurality of free-text keystrokes entered by a respective user on a computer that is part of the computing system, and storing each user profile in the memory, subsequently collecting the keystrokes of a user to be authenticated as the user enters text on a keyboard connected to the computing system, creating a plurality of graphs based on the collection of keystrokes entered by the user and calculating n instance based tail area density (ITAD) metric, and then combining the ITAD metric for each graph duration into a single similarity score.