Dual-Classifier Access Control for Adaptive User Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing access control methods for user devices are inadequate as they provide persistent access once gained, lacking effective mechanisms to continuously authenticate users and adapt to changing behavior, which can lead to unauthorized access.
Innovation Solution
A computer-implemented method using machine learning classifiers to determine user behavior, employing a dual-classifier system where non-compliant behavior triggers credential-based authentication and continuous training to improve classification accuracy, ensuring dynamic and adaptive access control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional access control methods are used, then access is granted once authenticated, but security is weakened because access remains persistent without continuous verification
Solution Approach 1:
The access control system transitions from a static authentication model to a dynamic one by continuously monitoring user behavior through machine learning classifiers. The system adapts its security decisions in real-time based on behavioral patterns, making the access control mechanism dynamic rather than persistent, thereby improving security without requiring complete system redesign
Solution Approach 2:
The system implements continuous feedback loops where user behavior is constantly monitored, classified by machine learning models, and used to adjust access decisions. This feedback mechanism enables the system to verify user identity continuously based on behavioral patterns, resolving the contradiction between maintaining simple access control and improving security reliability
2Measurement precision
If a single machine learning classifier is used for behavior classification, then the system is simple, but accuracy decreases when encountering non-compliant behavior patterns
Solution Approach 1:
The classification system is segmented into multiple specialized classifiers: a first classifier for compliant behavior patterns and a second classifier for non-compliant behavior patterns. This segmentation allows each classifier to specialize in specific behavior types, improving overall classification accuracy while managing complexity through modular design
Solution Approach 2:
The system changes the operational parameters of the classification process by switching between different classifiers based on the compliance status of detected behavior. When compliant behavior is detected, the first classifier is used; when non-compliant behavior is detected, the second classifier is activated, optimizing accuracy for different behavior regimes
3Reliability
If continuous behavioral monitoring is implemented, then security is improved through adaptive access control, but computational resources and system complexity increase
Solution Approach 1:
The system applies partial monitoring action by focusing computational resources on classifying only the most relevant behavioral features rather than continuously processing all possible data streams. This selective approach maintains continuous authentication reliability while reducing overall computational energy consumption
Data Source
AI summary
A computer implemented method of access control for a user device having at least one component for determining behaviors of the user. The method including accessing a first machine learning classifier trained based on at least one prior behavior of the user using the device, the classifier classifying user behavior as compliant or non-compliant. The method further including, in response to a determination that a subsequent behavior is classified as non-compliant, accessing a second machine learning classifier trained based on at least one prior behavior of the user using the device where the prior behavior is classified as non-compliant by the first classifier. The method further including, in response to a determination that the subsequent behavior is classified as non-compliant by the second classifier, requesting a credential-based authentication of the user and constructively training one of the machine learning classifiers based on the credential-based authentication result.