Edge ML Classifier for Request Pre-screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Authentication journeys are resource-intensive and vulnerable to malicious and probing requests, which can lead to distributed denial of service attacks, as they involve multiple steps and risk analysis from various sources, making them targets for bot detection and intrusion attempts.
Innovation Solution
Implementing machine learning classifiers at the network edge to pre-screen incoming requests, using signals such as user ID, geographical location, device type, and request headers to estimate risk and channel requests, thereby reducing the need for full authentication journeys and mitigating bot activity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If authentication journeys are implemented with multiple steps and risk analysis from various sources, then authentication security is improved, but resource consumption increases and vulnerability to denial of service attacks worsens
Solution Approach 1:
The patent implements a pre-screening mechanism that analyzes requests before they enter the full authentication journey. By performing preliminary risk assessment using machine learning classifiers at the edge of the network, the system identifies and blocks worthless requests early, preventing them from consuming resources during the complete authentication process while maintaining security for legitimate requests
Solution Approach 2:
The authentication process is segmented into distinct stages: pre-screening at the network edge using lightweight machine learning models, and full authentication journeys for requests that pass pre-screening. This segmentation allows the system to apply different levels of scrutiny appropriately, reducing overall resource consumption while maintaining security
2Loss of energy
If pre-screening mechanisms are implemented at the network edge, then resource consumption is reduced, but request processing time may increase
Solution Approach 1:
The patent replaces traditional mechanical filtering mechanisms with machine learning-based pre-screening. The machine learning classifiers are trained to rapidly evaluate request characteristics and predict whether requests are worthwhile or worthless, enabling fast decision-making that reduces both resource consumption and processing time compared to conventional approaches
Data Source
AI summary
The technology screens malicious and probing requests to establish an authorized session, collectively worthless requests, directed to a protected application, by using an identity gateway (IG) positioned as a network edge component on a network. Screening occurs before starting an authentication or authorization journey or other resource consuming interaction with a protected application. The screening setup involves provisioning a ML classifier at an edge server accessible by the IG. When the request to establish an authorized session is received by the IG, screening involves the ML classifier accepting features and outputting a score predicting whether the request is worthless. The IG compares the score to a threshold. Based on the score, the IG may determine to limit the worthless request at the network edge so that the request does not invoke the authorization journey or perform the resource consuming interaction.