Score Transformation Table for Access Control Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource security systems face challenges in maintaining consistent scoring over time due to changing access request behaviors and scoring model updates, leading to inconsistent risk level assessments and potential unauthorized access or denial of legitimate users.
Innovation Solution
A score transformation table is generated to dynamically adapt to changes in the distribution of scores, mapping raw scores from a scoring model to abbreviated scores that consistently correspond to certain risk levels, ensuring predictability and consistency in access request outcomes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a risk scoring model is used to identify fraudulent access requests, then unauthorized access is prevented, but authorized users may be rejected due to score inconsistencies over time
Solution Approach 1:
The patent transforms the raw risk score parameter into a percentile ranking parameter. Instead of using absolute score values that may drift over time, the system calculates the percentile rank of each score within a rolling time window, ensuring that the meaning of each score remains consistent (e.g., top 10% always represents the highest risk) even as scoring models are updated or behaviors change.
Solution Approach 2:
The system implements feedback by periodically recalculating percentile rankings based on historical score distributions and using this information to adjust current access decisions. The rolling time window provides continuous feedback about score distribution changes, allowing the system to adapt to new fraud patterns while maintaining consistent risk assessment standards.
2Adaptability or versatility
If scoring models are periodically updated to adapt to changing fraud patterns, then detection accuracy improves, but score distribution consistency deteriorates
Solution Approach 1:
The patent introduces dynamics by implementing a rolling time window that continuously adapts to new score distributions. Instead of using a fixed historical period, the window rolls forward in time, automatically incorporating new data while excluding older data. This allows the system to adapt to changing fraud patterns through model updates while maintaining score distribution stability through continuous recalibration.
Solution Approach 2:
The system performs periodic recalibration of percentile rankings at defined intervals or triggers. This periodic action ensures that score distributions are regularly normalized, preventing drift accumulation while allowing legitimate adaptations to new fraud patterns between recalibration cycles. The periodic normalization maintains stability without preventing adaptability.
3Object-affected harmful factors
If score thresholds are set to reject high-risk requests, then fraudulent access is blocked, but legitimate requests may be denied due to score drift
Solution Approach 1:
The patent changes the threshold parameter from an absolute score value to a percentile-based threshold. Instead of rejecting requests with scores above a fixed value (e.g., 80), the system rejects requests in the top percentile (e.g., top 10%). This ensures that the threshold always corresponds to the same relative risk level, preventing legitimate users from being denied access due to score drift while maintaining effective fraud blocking.
Data Source
AI summary
Resources can be secured by a resource security system. The resource security system can determine whether to grant or deny access to resources using authorization information in an access request. The resource security system can also determine whether the access request is legitimate or fraudulent using risk scoring models. A score transformation table can be used to provide consistency in the risk level for a particular score over time. The score transformation table can be based on a target score profile and a precision format (e.g., integer or floating point). The score transformation table can dynamically adapt based on the trending top percent of risk and can account for changes in the distribution of scores over time or by weekday. The scores can be used to determine an access request outcome. Access to the resource can be accepted or rejected based on the outcome.