Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Principal Component Analysis (PCA) is a commonly used unsupervised exploratory analysis technique. It is also frequently used in dimensionality reduction. This preliminary paper investigates the feasibility of three variants of PCA, i.e. independent PCA (iPCA), sparse PCA (sPCA), and sparse independent PCA (siPCA) on forensic classification of paper based on their IR spectral data. After that, Linear Discriminant Analysis (LDA) models were built using the Principal Components (PCs) produced by the PCA and the three aforementioned variants. The performances of all these four LDA models, i.e. PCA-DA, iPCA-DA, sPCA-DA and siPCA-DA, were evaluated via leave-one-out cross-validation on the data set. The results obtained show that iPCA-DA and siPCA-DA are the most effective models with 100.0% classification accuracy. Then, the effectiveness of siPCA and iPCA models was evaluated based on posterior probability used for predictions of class membership that were derived from leave-one-out cross-validation. As a conclusion, siPCA is identified as the best classification model.

Original languageEnglish
Title of host publicationAdvances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015
PublisherAmerican Institute of Physics Inc.
Volume1750
ISBN (Electronic)9780735414075
DOIs
Publication statusPublished - 21 Jun 2016
Event23rd Malaysian National Symposium of Mathematical Sciences: Advances in Industrial and Applied Mathematics, SKSM 2015 - Johor Bahru, Malaysia
Duration: 24 Nov 201526 Nov 2015

Other

Other23rd Malaysian National Symposium of Mathematical Sciences: Advances in Industrial and Applied Mathematics, SKSM 2015
CountryMalaysia
CityJohor Bahru
Period24/11/1526/11/15

Fingerprint

principal components analysis
predictions

Keywords

  • independent PCA
  • IR spectrum
  • PCA
  • sparse independent PCA
  • sparse PCA

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Lee, L. C., Liong, C. Y., Osman, K., & Jemain, A. A. (2016). Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum. In Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015 (Vol. 1750). [060012] American Institute of Physics Inc.. https://doi.org/10.1063/1.4954617

Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum. / Lee, Loong Chuen; Liong, Choong Yeun; Osman, Khairul; Jemain, Abdul Aziz.

Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015. Vol. 1750 American Institute of Physics Inc., 2016. 060012.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lee, LC, Liong, CY, Osman, K & Jemain, AA 2016, Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum. in Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015. vol. 1750, 060012, American Institute of Physics Inc., 23rd Malaysian National Symposium of Mathematical Sciences: Advances in Industrial and Applied Mathematics, SKSM 2015, Johor Bahru, Malaysia, 24/11/15. https://doi.org/10.1063/1.4954617
Lee LC, Liong CY, Osman K, Jemain AA. Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum. In Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015. Vol. 1750. American Institute of Physics Inc. 2016. 060012 https://doi.org/10.1063/1.4954617
Lee, Loong Chuen ; Liong, Choong Yeun ; Osman, Khairul ; Jemain, Abdul Aziz. / Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum. Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015. Vol. 1750 American Institute of Physics Inc., 2016.
@inproceedings{e803001240c440db9502e0625ec141fa,
title = "Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum",
abstract = "Principal Component Analysis (PCA) is a commonly used unsupervised exploratory analysis technique. It is also frequently used in dimensionality reduction. This preliminary paper investigates the feasibility of three variants of PCA, i.e. independent PCA (iPCA), sparse PCA (sPCA), and sparse independent PCA (siPCA) on forensic classification of paper based on their IR spectral data. After that, Linear Discriminant Analysis (LDA) models were built using the Principal Components (PCs) produced by the PCA and the three aforementioned variants. The performances of all these four LDA models, i.e. PCA-DA, iPCA-DA, sPCA-DA and siPCA-DA, were evaluated via leave-one-out cross-validation on the data set. The results obtained show that iPCA-DA and siPCA-DA are the most effective models with 100.0{\%} classification accuracy. Then, the effectiveness of siPCA and iPCA models was evaluated based on posterior probability used for predictions of class membership that were derived from leave-one-out cross-validation. As a conclusion, siPCA is identified as the best classification model.",
keywords = "independent PCA, IR spectrum, PCA, sparse independent PCA, sparse PCA",
author = "Lee, {Loong Chuen} and Liong, {Choong Yeun} and Khairul Osman and Jemain, {Abdul Aziz}",
year = "2016",
month = "6",
day = "21",
doi = "10.1063/1.4954617",
language = "English",
volume = "1750",
booktitle = "Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015",
publisher = "American Institute of Physics Inc.",

}

TY - GEN

T1 - Comparison of several variants of principal component analysis (PCA) on forensic analysis of paper based on IR spectrum

AU - Lee, Loong Chuen

AU - Liong, Choong Yeun

AU - Osman, Khairul

AU - Jemain, Abdul Aziz

PY - 2016/6/21

Y1 - 2016/6/21

N2 - Principal Component Analysis (PCA) is a commonly used unsupervised exploratory analysis technique. It is also frequently used in dimensionality reduction. This preliminary paper investigates the feasibility of three variants of PCA, i.e. independent PCA (iPCA), sparse PCA (sPCA), and sparse independent PCA (siPCA) on forensic classification of paper based on their IR spectral data. After that, Linear Discriminant Analysis (LDA) models were built using the Principal Components (PCs) produced by the PCA and the three aforementioned variants. The performances of all these four LDA models, i.e. PCA-DA, iPCA-DA, sPCA-DA and siPCA-DA, were evaluated via leave-one-out cross-validation on the data set. The results obtained show that iPCA-DA and siPCA-DA are the most effective models with 100.0% classification accuracy. Then, the effectiveness of siPCA and iPCA models was evaluated based on posterior probability used for predictions of class membership that were derived from leave-one-out cross-validation. As a conclusion, siPCA is identified as the best classification model.

AB - Principal Component Analysis (PCA) is a commonly used unsupervised exploratory analysis technique. It is also frequently used in dimensionality reduction. This preliminary paper investigates the feasibility of three variants of PCA, i.e. independent PCA (iPCA), sparse PCA (sPCA), and sparse independent PCA (siPCA) on forensic classification of paper based on their IR spectral data. After that, Linear Discriminant Analysis (LDA) models were built using the Principal Components (PCs) produced by the PCA and the three aforementioned variants. The performances of all these four LDA models, i.e. PCA-DA, iPCA-DA, sPCA-DA and siPCA-DA, were evaluated via leave-one-out cross-validation on the data set. The results obtained show that iPCA-DA and siPCA-DA are the most effective models with 100.0% classification accuracy. Then, the effectiveness of siPCA and iPCA models was evaluated based on posterior probability used for predictions of class membership that were derived from leave-one-out cross-validation. As a conclusion, siPCA is identified as the best classification model.

KW - independent PCA

KW - IR spectrum

KW - PCA

KW - sparse independent PCA

KW - sparse PCA

UR - http://www.scopus.com/inward/record.url?scp=84984576605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84984576605&partnerID=8YFLogxK

U2 - 10.1063/1.4954617

DO - 10.1063/1.4954617

M3 - Conference contribution

AN - SCOPUS:84984576605

VL - 1750

BT - Advances in Industrial and Applied Mathematics: Proceedings of 23rd Malaysian National Symposium of Mathematical Sciences, SKSM 2015

PB - American Institute of Physics Inc.

ER -