Historical Risk Assessment for Online Access Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing risk prediction models for online access control primarily provide a predicted risk score without detailing the impact of each attribute on the risk score change over time, limiting their effectiveness in risk mitigation.
Innovation Solution
The system performs an automated historical risk assessment by analyzing the impact of each attribute on risk indicator changes over time, aggregating these impacts to identify key attributes affecting the risk score, and transmitting these results for improved risk management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If risk prediction models provide only a predicted risk score, then the model complexity is low and computation is fast, but the ability to analyze attribute impacts and support risk mitigation is insufficient
Solution Approach 1:
The patent segments the risk assessment process into two distinct components: (1) a risk prediction model that generates risk scores, and (2) an impact analysis model that separately evaluates the contribution of each attribute to risk changes. This segmentation allows the system to provide detailed attribute impact information without requiring the entire system to be fundamentally more complex, as each component has a specialized function.
Solution Approach 2:
The patent introduces an intermediary impact analysis model that acts as a mediator between the risk prediction model and the user. This intermediary component translates the complex internal workings of the risk prediction model into interpretable attribute impact information, allowing users to understand which attributes contribute to risk changes without needing to comprehend the underlying complex model architecture.
2Reliability
If the system analyzes the impact of each attribute on risk score changes over time, then the effectiveness of risk mitigation is improved, but the computational time and processing resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-processing and storing attribute impact information as the risk prediction model is trained and updated. The impact analysis model is prepared in advance with the necessary computational frameworks (such as integrated gradients) to quickly evaluate attribute contributions when risk assessments are requested, reducing computational time during actual risk evaluation.
Solution Approach 2:
The patent replaces traditional mechanical re-computation of attribute impacts with a more efficient computational approach using integrated gradients and other approximation methods. Instead of re-running the entire risk prediction model to assess attribute impacts, the system uses substituted computational techniques that provide accurate impact information with significantly reduced processing time.
3Ease of operation
If the system provides detailed assessment results including ordered lists of attributes by total impact, then the usefulness for risk mitigation is improved, but the data processing and output generation complexity increases
Solution Approach 1:
The patent implements feedback mechanisms where the impact analysis model continuously evaluates and ranks attributes based on their contribution to risk changes. This feedback loop processes attribute impact data and generates ordered lists that provide actionable insights for risk mitigation. The systematic feedback structure transforms complex processing requirements into organized, interpretable output that directly supports risk management decisions.
Solution Approach 2:
The patent changes the parameter representation of risk assessment results by transforming raw attribute impact values into ranked orders and relative contribution metrics. Instead of presenting complex multidimensional attribute data, the system reparameterizes the output as ordered lists showing which attributes have the greatest impact, making the information more usable for risk mitigation while managing processing complexity through intelligent data transformation.
Data Source
AI summary
Systems and methods for automated historical risk assessment for risk mitigation in online access control are provided. An entity assessment server can receive a request to assess a risk indicator change from a first risk indicator to a second risk indicator. For each attribute used to generate the first risk indicator and second risk indicator, a first impact can be determined for changing from the first risk indicator to a third risk indicator between the first risk indicator and the second risk indicator. A second impact similarly can be determined for changing from the third risk indicator to the second risk indicator. Aggregating the first impact and the second impact can determine a total impact of each attribute. Assessment results can be generated to include a list of attributes ordered according to the respective total impact and transmitted to a remote computing device for use in improving the risk indicator.