Higher-Order Anomaly Score for Detection Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly detection systems in various fields, such as computing and seismological systems, often fail to detect anomalies in a timely manner due to the complexity and volume of data generated, leading to potential undesirable consequences like security breaches or system failures.
Innovation Solution
The use of higher-order anomaly scores, which consider the relatedness of multiple anomaly detectors over time, to detect anomalies that may not be identified by first-order scores alone, allowing for more accurate and timely detection and remediation of issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple anomaly detectors are used to analyze large data sets, then detection accuracy is improved, but computing resource consumption increases
Solution Approach 1:
The patent segments the anomaly detection process into multiple independent first-order anomaly detectors that each analyze specific aspects of the data. These detectors are then combined through a higher-order anomaly score that synthesizes their results. This segmentation allows parallel processing of data subsets, improving detection accuracy while managing computing resources through modular, independent analysis units rather than a single resource-intensive comprehensive analyzer.
Solution Approach 2:
The patent introduces a higher-order dimension to anomaly detection by combining first-order anomaly scores from multiple detectors into a higher-order anomaly score. This dimensional transformation allows the system to capture complex anomaly patterns that individual detectors miss, improving detection accuracy without requiring each individual detector to process the entire data set at full complexity, thus managing computing resource consumption more efficiently.
2Loss of time
If traditional anomaly detection methods are used, then computing resources are conserved, but anomaly detection timeliness deteriorates
Solution Approach 1:
The patent implements preliminary action by having multiple first-order anomaly detectors pre-process and analyze different aspects of the data in parallel before combining their results into a higher-order anomaly score. This preliminary parallel processing accelerates anomaly detection timeliness by avoiding sequential analysis, while the modular structure manages system complexity through standardized, reusable detector components that can be independently developed and maintained.
Solution Approach 2:
The patent merges the results from multiple first-order anomaly detectors into a single higher-order anomaly score through a standardized combination mechanism. This merging approach improves detection timeliness by synthesizing parallel analysis results simultaneously, while managing system complexity through a unified integration framework that standardizes how individual detectors contribute to the overall anomaly assessment.
3Reliability
If first-order anomaly scores are used alone, then computing resources are reduced, but anomaly detection completeness deteriorates
Solution Approach 1:
The patent introduces a higher-order anomaly score as an intermediary that mediates between multiple first-order anomaly detectors and the final anomaly detection decision. This intermediary synthesizes the results from multiple specialized detectors, improving detection completeness by capturing anomalies that individual detectors miss, while managing methodology complexity through a standardized mediation layer that provides a clear interface between individual detectors and the overall detection system.
Data Source
AI summary
First-order anomaly scores are received from related anomaly detectors. Each first-order anomaly score indicates a likelihood of an anomaly at a target system. A relatedness measure of the related anomaly detectors is determined, based on the first-order anomaly scores that have been received. A higher-order anomaly score is determined based on the relatedness measure that has been determined. The higher-order anomaly score indicates a likelihood of an anomaly at the target system. An anomaly at the target system is detected based on the higher-order anomaly score.