Visualizing Multi-Dimensional Data Correlations

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Analyzing data for correlations is a time-consuming and processor-intensive task, especially when dealing with large datasets, and standard statistical tests often fail to discover correlations between attributes of different tables or data sets, particularly partial correlations.

Innovation Solution

A system for visualizing correlations between attributes in a dataset or across multiple datasets using graphical representations, where users can assign graphical elements like colors or shadings to attribute value ranges, allowing for the 'drag and drop' interaction to overlay graphical elements from one attribute onto another, facilitating the visual identification of potential correlations without requiring computationally expensive calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If standard statistical tests are used to discover correlations, then correlation thresholds can be established, but the tests fail to discover partial correlations and require manual threshold selection for each application domain

Engineering Contradiction:

Improvecorrelation detection accuracyVSAvoidapplicability across different domains

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent transforms the correlation detection problem from numerical parameter comparison to visual parameter representation. By converting attribute values into visual encodings (colors, sizes, shapes) and displaying them in parallel coordinate systems, the system enables correlation detection through visual pattern recognition rather than statistical threshold comparison, thereby achieving both precision and cross-domain adaptability

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical statistical testing process with a visual information processing system. Instead of applying statistical formulas and algorithms to compute correlation coefficients, the system uses visual encoding and graphical display to allow human visual processing to detect correlations, effectively substituting computational mechanics with visual perception mechanics

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If correlation calculations are performed on large data sets with thousands or millions of records, then comprehensive correlation analysis can be conducted, but the process becomes time consuming and processor intensive

Engineering Contradiction:

Improvecompleteness of correlation analysisVSAvoidtime required for correlation calculations

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts the essential correlation information from large data sets by transforming attribute values into visual representations. Instead of performing exhaustive calculations on all records, the system extracts key patterns through visual encoding, allowing users to identify correlations by observing visual patterns in the graphical display rather than waiting for complete computational analysis

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent creates visual copies of attribute data in the form of encoded graphical elements. By representing numerical attribute values as visual properties (colors, sizes, positions) in parallel coordinate systems, the system allows users to perceive correlation patterns directly from these visual copies without performing actual correlation calculations on the original numerical data

Inventive Principle:

Principle #26Copying

Data Source

PatentUS8928663B2Visualizing correlations in multi-dimensional data

Publication Date: 2015.01.06 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

A system for visualizing correlations between attributes in a data set or across multiple data sets is provided. A user may view a graphical representation (e.g., a histogram) of attribute values for a first attribute. The user may assign a variety of graphical indicators to various value ranges of the first attribute. The user may view a graphical representation of the second attributes. The user may “drag and drop” the graphical representation of the first attributes onto the graphical representation of the second attributes. The graphical representation of the second attributes may be updated to incorporate the graphical elements assigned by the user to the value ranges of the first attribute. The user may visually see potential correlations between the first and the second attributes based on the graphical elements associated with the first attributes displayed with the associated second attributes.