Landmark Point Selection for Interactive Data Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing large multidimensional datasets are inefficient and fail to identify important relationships, often breaking relationships and being too sensitive to large scale distances, requiring sophisticated experts and non-interactive visualizations that do not allow for exploratory data analysis.
Innovation Solution
A method involving the selection of landmark points, where data points are grouped by calculating distances to initial landmark points, expanding the set with the longest distances, and generating an interactive visualization of nodes and edges to represent the data, allowing for exploratory analysis and improved relationship identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If clustering methods are used to process large datasets, then data processing capability is improved, but important relationships are broken and analysis precision deteriorates
Solution Approach 1:
The patent divides the large dataset into multiple clusters, where each cluster is represented by a centroid. This segmentation allows efficient processing of large datasets while preserving local relationships within each cluster. The centroids serve as representative points that maintain the essential characteristics of their respective clusters without breaking important relationships.
Solution Approach 2:
The patent introduces centroids as intermediary points between individual data points and the overall dataset structure. These centroids act as mediators that preserve relationships by representing groups of points while enabling efficient computation. The distance metrics are calculated between centroids rather than all individual points, maintaining relationship integrity while improving processing efficiency.
2Speed
If linear algebraic and analytic methods are used, then analysis speed is improved, but detail is lost due to sensitivity to large scale distances
Solution Approach 1:
The patent applies local quality by using distance metrics that are sensitive to local variations within clusters while being robust to large-scale distances between clusters. The relationship preservation technique focuses on maintaining detailed local relationships within each cluster rather than treating all distances uniformly, thereby preserving detail while enabling fast analysis through cluster-based processing.
3Loss of information
If traditional graph visualization methods are used, then relationship depiction is improved, but interactivity and exploratory analysis capability deteriorate
Solution Approach 1:
The patent implements dynamic visualization where the graph structure and relationships can be interactively explored and modified. The system allows users to dynamically adjust parameters, explore different cluster configurations, and perform exploratory data analysis on the visualized relationships. This dynamic approach maintains high-quality relationship depiction while enabling interactivity and user exploration.
Data Source
AI summary
An exemplary method comprises receiving data points, selecting a first subset of the data points to generate an initial set of landmarks, each data point of the first subset defining a landmark point and for each non-landmark data point: calculating first data point distances between a respective non-landmark data point and each landmark point of the initial set of landmarks, identifying a first shortest data point distance from among the first data point distances between the respective non-landmark data point and each landmark point of the initial set of landmarks, and storing the first shortest data point distance as a first landmark distance for the respective non-landmark data point. The method further comprising identifying a non-landmark data point with a longest first landmark distance in comparison with other first landmark distances and adding the identified non-landmark data point associated as a first landmark point to the initial set of landmarks.