Cluster-Based Risk Assessment for Accurate Access Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning models fail to accurately facilitate risk assessment operations, particularly in determining whether a requesting entity poses a risk to a provider, which can lead to unauthorized interactions or resource loss.
Innovation Solution
Employing clustering operations, specifically high-dimensional clustering, to identify a target cluster and its nearest neighbors, allowing for the prediction of future risk assessment values based on historical data, thereby improving access control to computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional machine learning models are used for risk assessment, then predictions can be provided, but the models fail to accurately facilitate risk assessment operations and cannot determine whether a requesting entity poses a risk
Solution Approach 1:
The patent segments the risk assessment process into multiple components: historical data retrieval, cluster identification, nearest neighbor selection, and risk score calculation. This segmentation allows each component to be optimized independently, improving overall assessment accuracy while maintaining reliability through structured multi-step evaluation
Solution Approach 2:
The system performs preliminary actions by pre-processing historical risk assessment data into clustered groups before actual risk assessment is needed. This preliminary clustering enables faster and more accurate real-time risk determination by comparing new requests against pre-organized historical patterns
2Measurement precision
If high dimensional clustering is used to identify target clusters and nearest neighbors, then risk prediction accuracy is improved, but computational and storage requirements increase
Solution Approach 1:
The patent applies partial action by selecting only the most relevant features for clustering and limiting the number of nearest neighbor clusters considered. This partial processing maintains high prediction accuracy while reducing computational complexity and storage requirements compared to using all available dimensions
Solution Approach 2:
The system changes parameters by transforming raw historical data into clustered representations with reduced dimensionality. This parameter transformation preserves essential risk patterns while significantly reducing the computational and storage burden of handling high-dimensional data
Data Source
AI summary
Systems and methods for predicting future risk for a target entity are provided. A risk assessment system receives historical risk assessment data of the target entity and identifies a target cluster that matches the historical risk assessment data. The target cluster is identified from a group of clusters determined using high dimensional clustering based on risk assessment data of a set of entities. The risk assessment system identifies a set of nearest neighbors of the target cluster and determines a prediction of future risk for the target entity based on the target cluster and the set of nearest neighbors. The risk assessment system transmits a responsive message, which can include the prediction of future risk, to a remote computing device for use in controlling access of the target entity to one or more interactive computing environments.