Date of Award
12-2005
Level of Access Assigned by Author
Campus-Only Thesis
Degree Name
Master of Science (MS)
Department
Electrical and Computer Engineering
Advisor
Mohamad T. Musavi
Second Committee Member
Habtom Ressom
Third Committee Member
Sean Ireland
Abstract
Principal Components Analysis (PCA) is a conventional linear technique for projecting multidimensional data onto lower dimensional spaces with minimal loss of variance. However, there are several applications where the data is not linear; in these cases linear PCA is not the optimal method to recover this subspace and thus account for the largest proportion of variance in the data. In this thesis, a non-linear PCA (IVLPCA) method is developed using a new technique that combines Radial Basis function with Particle Swarm optimization. The new technique is evaluated and compared to other standard methods in the applications of function approximation, feature extraction, and process monitoring.
Recommended Citation
Driss, Mohamed Naoufel, "Non-linear PCA Based on Radial Basis Function and Particle Swarm Optimization" (2005). Electronic Theses and Dissertations. 986.
https://digitalcommons.library.umaine.edu/etd/986