Probabilistic kernel principal component analysis through time

Mauricio Alvarez; Ricardo Henao

doi:10.1007/11893028_83

Probabilistic kernel principal component analysis through time

Mauricio Alvarez, Ricardo Henao

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

This paper introduces a temporal version of Probabilistic Kernel Principal Component Analysis by using a hidden Markov model in order to obtain optimized representations of observed data through time. Recently introduced. Probabilistic Kernel Principal Component Analysis overcomes the two main disadvantages of standard Principal Component Analysis, namely, absence of probability density model and lack of high-order statistical information due to its linear structure. We extend this probabilistic approach of KPCA to mixture models in time, to enhance the capabilities of transformation and reduction of time series vectors. Results over voice disorder databases show improvements in classification accuracies even with highly reduced representations. © Springer-Verlag Berlin Heidelberg 2006.

Original language	English (US)
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer Verlag
Pages	747-754
Number of pages	8
ISBN (Print)	3540464794
DOIs	https://doi.org/10.1007/11893028_83
State	Published - Jan 1 2006
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2023-02-15

Access to Document

10.1007/11893028_83

Cite this

@inproceedings{387bee7292d642d8ae2dec9976cd51c5,

title = "Probabilistic kernel principal component analysis through time",

abstract = "This paper introduces a temporal version of Probabilistic Kernel Principal Component Analysis by using a hidden Markov model in order to obtain optimized representations of observed data through time. Recently introduced. Probabilistic Kernel Principal Component Analysis overcomes the two main disadvantages of standard Principal Component Analysis, namely, absence of probability density model and lack of high-order statistical information due to its linear structure. We extend this probabilistic approach of KPCA to mixture models in time, to enhance the capabilities of transformation and reduction of time series vectors. Results over voice disorder databases show improvements in classification accuracies even with highly reduced representations. {\textcopyright} Springer-Verlag Berlin Heidelberg 2006.",

author = "Mauricio Alvarez and Ricardo Henao",

note = "Generated from Scopus record by KAUST IRTS on 2023-02-15",

year = "2006",

month = jan,

day = "1",

doi = "10.1007/11893028_83",

language = "English (US)",

isbn = "3540464794",

pages = "747--754",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

address = "Germany",

}

TY - GEN

T1 - Probabilistic kernel principal component analysis through time

AU - Alvarez, Mauricio

AU - Henao, Ricardo

N1 - Generated from Scopus record by KAUST IRTS on 2023-02-15

PY - 2006/1/1

Y1 - 2006/1/1

N2 - This paper introduces a temporal version of Probabilistic Kernel Principal Component Analysis by using a hidden Markov model in order to obtain optimized representations of observed data through time. Recently introduced. Probabilistic Kernel Principal Component Analysis overcomes the two main disadvantages of standard Principal Component Analysis, namely, absence of probability density model and lack of high-order statistical information due to its linear structure. We extend this probabilistic approach of KPCA to mixture models in time, to enhance the capabilities of transformation and reduction of time series vectors. Results over voice disorder databases show improvements in classification accuracies even with highly reduced representations. © Springer-Verlag Berlin Heidelberg 2006.

AB - This paper introduces a temporal version of Probabilistic Kernel Principal Component Analysis by using a hidden Markov model in order to obtain optimized representations of observed data through time. Recently introduced. Probabilistic Kernel Principal Component Analysis overcomes the two main disadvantages of standard Principal Component Analysis, namely, absence of probability density model and lack of high-order statistical information due to its linear structure. We extend this probabilistic approach of KPCA to mixture models in time, to enhance the capabilities of transformation and reduction of time series vectors. Results over voice disorder databases show improvements in classification accuracies even with highly reduced representations. © Springer-Verlag Berlin Heidelberg 2006.

UR - http://link.springer.com/10.1007/11893028_83

UR - http://www.scopus.com/inward/record.url?scp=33750603446&partnerID=8YFLogxK

U2 - 10.1007/11893028_83

DO - 10.1007/11893028_83

M3 - Conference contribution

SN - 3540464794

SP - 747

EP - 754

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -

Probabilistic kernel principal component analysis through time

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this