Navigation Anomaly Detection via Classification and Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems lack an effective anomaly detection mechanism, leading to errors and inefficiencies in providing location-based information services, which is critical in today's competitive market with growing consumer expectations and commercial pressures.
Innovation Solution
A navigation system incorporating a supervised learning module for determining a classification model, a clustering module for identifying clusters, and an evaluation module to detect anomalies based on the classification model and clusters, allowing for accurate display on devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If anomaly detection mechanism is added to navigation system, then reliability of navigation information is improved, but device complexity increases
Solution Approach 1:
The anomaly detection system is segmented into distinct functional modules: a classification module that determines category labels for points of interest, a clustering module that groups similar points together, and an anomaly detection module that identifies outliers. This segmentation allows each module to perform its specific function independently, making the overall system more manageable and easier to implement while maintaining high reliability.
Solution Approach 2:
The patent introduces intermediary data structures and processing steps between the raw navigation data and the final anomaly detection. Specifically, the classification and clustering modules serve as intermediaries that transform raw point of interest data into structured representations, which then feed into the anomaly detection algorithm. This intermediary layer simplifies the detection process and improves reliability by filtering out noise before anomaly identification.
2Measurement precision
If classification model and clustering algorithms are implemented, then accuracy of point of interest assessment is improved, but computation time increases
Solution Approach 1:
The system performs preliminary classification and clustering actions on point of interest data before anomaly detection is required. By pre-processing the data into classified categories and clustered groups, the system prepares the information in advance, which reduces the computation time needed during actual anomaly detection operations. This preliminary action allows for accurate assessment while minimizing real-time computational requirements.
Solution Approach 2:
The patent applies partial clustering by focusing only on relevant features and parameters when forming clusters, rather than analyzing all possible attributes. This selective clustering approach maintains sufficient accuracy for point of interest assessment while significantly reducing the computational burden. The system clusters based on the most informative features, achieving the necessary precision without excessive computation.
Data Source
AI summary
A method of operation of a navigation system includes: determining a classification model for a target field for assessing a point of interest; determining a cluster for the target field for assessing the point of interest; and determining an anomaly based on the classification model and the cluster for displaying on a device.