An Algorithm for the Analysis and Visualization of High Dimensional Space
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.
Focuses on the paths through high-dimensional space rather than space as a whole Simplifies N-dimensional path data sets into two-dimensional data
Data analysis Data visualization
Nathan DennyPurdue Research ComputingPurdue HubZero
美國
