 |
General: 663 |
P2: Nonlinear Dimensionality Reduction Applied to the Binary Classification of Images
Author: Chae Clark , Advisor: Kasso Okoudjou (MATH)
Problem Statement Presentation
Project Proposal
Abstract
In this project we are interested in the reduction of high-dimensional data points x from a space RD to a lower dimensional space Rd (where d << D) in a way that preserves certain important characteristics. In particular, we are interested in reducing the size of high-dimensional points for their application in the binary classification of signals. We rst examine a MatLab implementation of the Locally Linear Embeddings (LLE) algorithm, and it to a specific image database. We then use the output of the LLE and the original dataset to test and compare the performance ability of a MatLab implementation of the support vector machine.
MidYear Progress Report and
Presentation
Final Presentation ,
Final Report
