Is this project an undergraduate, graduate, or faculty project?
Undergraduate
individual
What campus are you from?
Daytona Beach
Authors' Class Standing
Kian Greene, Junior
Lead Presenter's Name
Kian Greene
Faculty Mentor Name
Mihhail Berezovski
Abstract
The objective of this Research Experience for Undergraduates is to compare features between a Principal Component Analysis (PCA) approach and a Non-negative Matrix Factorization (NNMF) approach towards interpreting a UV-vis spectral data set. Furthermore, the development of a predictive model to determine plutonium concentrations in different nitric acid solutions based on the PCA and NNMF information is also a goal of this project. At first, MATLAB software was used to read and work with the original spectral data set. However, as the project developed, Python was then used to implement both dimensionality reduction algorithms. Moreover, the linear models made from the decomposition outputs were also created in Python. Comparing the models shows that a PCA approach towards this spectral analysis case would be more efficient than the NNMF approach, but an NNMF model appears to fit the information better than the PCA model.
Did this research project receive funding support from the Office of Undergraduate Research.
Yes, Student Internal Grant
A Non-negative Matrix Factorization Approach to Computing “Fingerprints” in Spectra of Nuclear Materials
The objective of this Research Experience for Undergraduates is to compare features between a Principal Component Analysis (PCA) approach and a Non-negative Matrix Factorization (NNMF) approach towards interpreting a UV-vis spectral data set. Furthermore, the development of a predictive model to determine plutonium concentrations in different nitric acid solutions based on the PCA and NNMF information is also a goal of this project. At first, MATLAB software was used to read and work with the original spectral data set. However, as the project developed, Python was then used to implement both dimensionality reduction algorithms. Moreover, the linear models made from the decomposition outputs were also created in Python. Comparing the models shows that a PCA approach towards this spectral analysis case would be more efficient than the NNMF approach, but an NNMF model appears to fit the information better than the PCA model.