Patrick E. Hoffman Home Page

I'm no longer a student at UMass, but they let former Doctoral students keep web pages.

My thesis was: Table Visualizations, a Formal Model and its applications . This researched machine learning and visualization (Datamining). 

I’m currently teaching Electronics at The Benjamin Franklin Institute of Technology.

If you would like more information about my recent employment, check out my linked in profile


RADVIZ and other Visualization  tools are now available-see directory in this link


Read the readme.txt   - more documentation is coming!!


If you use these data mining and visualization  tools please acknowledge Umass Lowell and myself.

Consider this a a GNU license.


This includes 3 D Radviz  (RadVizX.exe) and Superviz ( a jar file sv020408.jar)

To run simply run RadVizX on a windows platform (ignore registry error).


In RadvizX open up the file carsdb.csv


Please let me know if you have tried these programs.  (see email at bottom)


For more information, please see the papers below. For more personal details see here
For more information about my research or consulting/employment opportunities please contact me at my e-mail address below.

If you are interested in running you might want to go here:
Travis Landreth Cross Country, Track  and Running Analysis
Go straight to the Best Track-Running Calculator

Besides running and running analysis, I spent a few years Juggling.

One of my research areas is Data mining and Visualization applied to Bio/Chem Informatics.

The data mining visualizations below were developed from my research at the IVPR and
along with some unique machine learning algorithms are potentially available to interested parties.


The idea of Dimensional Anchors can be used for creating an infinite number of information visualizations.  Some of these can be seen in these Grand Tour visualization movies of the Car and Iris datasets.

Iris Grand Tour Movie                       Car Dataset Grand Tour Movie

Here is the paper that introduces RadViz - A visualization that can effectively handle thousands of dimensions - DNA Visual And Analytic Data Mining presented at Visualization '97.

This paper extends the RadViz concept and shows how to produce any high dimensional visualization from a graphic primitive called a Dimensional Anchor Dimensional Anchors: A Graphic Primitive for Multidimensional Multivariate Information Visualizations

Two additional papers survey high dimensional visualizations and benchmark high dimensional visualizations

If you would like to e-mail me feel free

special link 

special link1

Patrick Hoffman           Institute for Visualization and Perception Research               University of Massachusetts-Lowell
                                  Computer Science Department
                                  One University Ave.         Lowell, MA 01854