About | Download | Examples | Docs | ApiDocs | Developer

HiSee is a tool for visualizing high-dimensional datasets. HiSee projects high-dimensional data to (currently) two dimensions using one of a variety of projection techniques. By comparing the way different projection algorithms present a dataset users can gain qualitative understanding of high-dimensional structures. Consider efforts to project a globe to two dimensions. Even though each projection introduces its own distortions, by comparing them one gets a sense of the three-dimensional structure of our world. HiSee makes use of the Piccolo zoomable user interface (ZUI), which allows users to emphasize and explore specific regions of a dataset. HiSee can be used as a standalone application or as a component in other programs (like Simbrain) which generate high dimensional data in real-time. Algorithms for fast addition of new datapoints are included. The program is written in Java and is open source. We are currently seeking developers to (among other things) contribute new projection algorithms and to implement a three-dimensional view.

Shown below is a bouquet of circles (in this case, a boquet of three circles) in a six dimensional space, under three different projections. Each projection illustrates successively more about the high-dimensional object.

HiSee and this site were created by Scott Hotton and Jeff Yoshimi.

SourceForge.net Logo