An Algorithm for the Analysis and Visualization of High Dimensional Space

Back to all technologies
Download as PDF
2014-DENN-66642
With the development of big data, visualization of data is seen as a key element in the overall process of extracting value from the 2.5 quintillion bytes of data that is created throughout the world each day. One such emerging area that could benefit from the analysis of this data is the Materials Genome Initiative, which was established to help businesses discover, develop, and deploy new materials at least twice as fast and at a fraction of the cost.

Researchers at Purdue University have developed an algorithm called MEANDER that produces two-dimensional data representations and graphic visualizations of paths that traverse distinct points in high-dimensional spaces. The resulting two-dimensional data can then be analyzed using a variety of planar data analysis programs that are incapable of working with high-dimensional space. This visualization can be both static images or animations and aid in the comprehension of patterns of behavior. MEANDER can be used in visualizing SQL queries, web and semantic searches, etc. This functionality is of growing importance as governmental and commercial entities emphasize informatics.

Advantages:
-Focuses on the paths through high-dimensional space rather than space as a whole
-Simplifies N-dimensional path data sets into two-dimensional data

Potential Applications:
-Data analysis
-Data visualization
(No issued patents found)
Purdue Office of Technology Commercialization
1801 Newman Road
West Lafayette, IN 47906

Phone: (765) 588-3475
Fax: (765) 463-3486
Email: otcip@prf.org