Multidimensional Scaling
Design:
Detailed Time Line and Tasks
Specification Apr 18-May 27
6 weeks
- Determine algorithm(s) to aquire 3D data.
- Determine interface requriments.
- Determine output requriments.
- Begin studying DirectX
Design May 30-July 1
5 weeks
- Design objects needed from requirments.
- Create wiki pages containing objcets, methods, members and algorithms.
Implementaiton July 5-Jul 29
4 weeks
- Implement objects.
- Build the interface.
- Testing
Testing Aug 01-Aug 12
2 weeks
- Test to ensure complete, correct product.
Timeline modifications
- 2 weeks added to spec on 2005-05-23
- 1 weeks removed from design on 2005-05-23
- 1 weeks removed from testing on 2005-05-23
(huiming)
- 1 weeks added to design on 2005-06-17
- 1 weeks removed from implementation on 2005-06-17
(huiming 2005-07-10)
- Since the algorithm wouldn't be implemented, testing time merged into implementation.
- Testing 2 weeks removed
The Basic Concept of Multidimensional Scaling
The following is the basic explanation of Multidimensional Scaling-- the 3D graph.
Multidimensional scaling (MDS) allows you to visualize how near points are to each other for many kinds of distance or dissimilarity measures and can produce a representation of your data in a small number of dimensions. MDS does not require raw data, but only a matrix of pairwise distances or dissimilarities.
MDS takes as an input a matrix of interpoint distances and creates a configuration of points by some mathematical matrix manipulation. Ideally, those points are in two or three dimensions, and the Euclidean distances between them reproduce the original distance matrix. Thus, a scatter plot of the points created by MDS provides a visual representation of the original distances. See the attachment below.
Team Members
Useful Links
C# OpenGL:
http://csgl.sourceforge.net/
An example:
http://www.exocortex.org/3dengine/
Direct3D:
http://www.drunkenhyena.com/docs/d3d_tutorial.php