Visualization of Multivariate Data

Web Published:


Novel Tool for Visualization of Multivariate Data


Princeton Docket # 13-2856-1

Researchers at Princeton University have developed novel visualization software to enable comparing and analyzing variables from large complex datasets.  Princeton is seeking a commercial partner to license the technology.

Representing trivariate and higher order data visually is challenging because multiple samples can overlap and obscure the visualization.  This new software overcomes this limitation and enables unambiguous display of data points with multiple variables.  This type of data frequently occurs in financial correlations, engineering analysis, and population studies.  Commercial applications, such as targeted marketing, can be improved by identifying population segments.   For example, the data for a large group of people could include their monthly salary, rent, and cell phone minutes.  It is useful to present the data graphically to see patterns and trends, for the purposes of product planning, policies, and other forms of analysis and decision making.

Scatter plots are commonly used but suffer when overlapping points obscure each other. Additionally, the order of plotting the points significantly affects the presentation of the data.  This new approach solves these problems by mapping the sample totals of trivariate data to the vertical axis of three-dimensional variable poles (VarPoles).  This enables representing all the samples in one diagram at the same time and eliminates the misrepresentation of data in scatter plots.

It is anticipated that this software could be used in many statistical fields, such as financial, engineering and population studies.

The U.S. Department of Energy's Princeton Plasma Physics Laboratory is a Collaborative National Center for plasma and fusion science. Its primary mission is to develop the scientific understanding and the key innovations which will lead to an attractive fusion energy source. Associated missions include conducting world-class research along the broad frontier of plasma science and providing the highest quality of scientific education. 


·         For Visualization of Multivariate Data



·         Novel

·         Deep and systemic

·         Highly sensitive


Intellectual Property and Commercialization Strategy

For further development and commercialization, Princeton University is pursuing a non-exclusive licensing strategy for the software.






Michael Tyerech

Princeton University Office of Technology Licensing 

(609) 258-6762



Patent Information:
For Information, Contact:
Michael Tyerech
former Princeton Sr. Licensing Associate
Princeton University
Eliot Feibush