LFDA model for the assessment of water quality through Microtox® using excitation-emission matrices

Oscar Martinez; Ranga Dabarera; Kamal Premaratne; Miroslav Kubat; James Douglas Englehardt

doi:10.3233/IDA-150291

LFDA model for the assessment of water quality through Microtox® using excitation-emission matrices

Oscar Martinez, Ranga Dabarera, Kamal Premaratne, Miroslav Kubat, James Douglas Englehardt

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Fluorescence spectroscopic excitation-emission matrices (EEMs) can be used to characterize dissolved organic matter in water with high sensitivity. Our goal is to predict the standard biosensor-based measurement Microtox® utilizing EEMs, pH, turbidity and conductivity, among other variables. EEMs have been modeled using novel latent fluorescent Dirichlet allocation (LFDA) based probabilistic graphical model. We found that nonparametric techniques offer a better mapping from, LFDA based EEMs scores and other measurements, to Microtox® measurements. The final decision on Microtox® measurement is given using an evidence fusion mechanism. In general, the novel LFDA based graphical model can be utilized in analyzing two dimensional data.

Original language	English (US)
Pages (from-to)	181-203
Number of pages	23
Journal	Intelligent Data Analysis
Volume	21
Issue number	1
DOIs	https://doi.org/10.3233/IDA-150291
State	Published - 2017

Keywords

fluorescence spectroscopy
Latent Dirichlet allocation
nonparametric estimation

ASJC Scopus subject areas

Theoretical Computer Science
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.3233/IDA-150291

Fingerprint Dive into the research topics of 'LFDA model for the assessment of water quality through Microtox® using excitation-emission matrices'. Together they form a unique fingerprint.

Cite this

@article{6f9279407be44b2ab049961033402647,

title = "LFDA model for the assessment of water quality through Microtox{\textregistered} using excitation-emission matrices",

abstract = "Fluorescence spectroscopic excitation-emission matrices (EEMs) can be used to characterize dissolved organic matter in water with high sensitivity. Our goal is to predict the standard biosensor-based measurement Microtox{\textregistered} utilizing EEMs, pH, turbidity and conductivity, among other variables. EEMs have been modeled using novel latent fluorescent Dirichlet allocation (LFDA) based probabilistic graphical model. We found that nonparametric techniques offer a better mapping from, LFDA based EEMs scores and other measurements, to Microtox{\textregistered} measurements. The final decision on Microtox{\textregistered} measurement is given using an evidence fusion mechanism. In general, the novel LFDA based graphical model can be utilized in analyzing two dimensional data.",

keywords = "fluorescence spectroscopy, Latent Dirichlet allocation, nonparametric estimation",

author = "Oscar Martinez and Ranga Dabarera and Kamal Premaratne and Miroslav Kubat and Englehardt, {James Douglas}",

year = "2017",

doi = "10.3233/IDA-150291",

language = "English (US)",

volume = "21",

pages = "181--203",

journal = "Intelligent Data Analysis",

issn = "1088-467X",

publisher = "IOS Press",

number = "1",

}

TY - JOUR

T1 - LFDA model for the assessment of water quality through Microtox® using excitation-emission matrices

AU - Martinez, Oscar

AU - Dabarera, Ranga

AU - Premaratne, Kamal

AU - Kubat, Miroslav

AU - Englehardt, James Douglas

PY - 2017

Y1 - 2017

N2 - Fluorescence spectroscopic excitation-emission matrices (EEMs) can be used to characterize dissolved organic matter in water with high sensitivity. Our goal is to predict the standard biosensor-based measurement Microtox® utilizing EEMs, pH, turbidity and conductivity, among other variables. EEMs have been modeled using novel latent fluorescent Dirichlet allocation (LFDA) based probabilistic graphical model. We found that nonparametric techniques offer a better mapping from, LFDA based EEMs scores and other measurements, to Microtox® measurements. The final decision on Microtox® measurement is given using an evidence fusion mechanism. In general, the novel LFDA based graphical model can be utilized in analyzing two dimensional data.

AB - Fluorescence spectroscopic excitation-emission matrices (EEMs) can be used to characterize dissolved organic matter in water with high sensitivity. Our goal is to predict the standard biosensor-based measurement Microtox® utilizing EEMs, pH, turbidity and conductivity, among other variables. EEMs have been modeled using novel latent fluorescent Dirichlet allocation (LFDA) based probabilistic graphical model. We found that nonparametric techniques offer a better mapping from, LFDA based EEMs scores and other measurements, to Microtox® measurements. The final decision on Microtox® measurement is given using an evidence fusion mechanism. In general, the novel LFDA based graphical model can be utilized in analyzing two dimensional data.

KW - fluorescence spectroscopy

KW - Latent Dirichlet allocation

KW - nonparametric estimation

UR - http://www.scopus.com/inward/record.url?scp=85009905282&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009905282&partnerID=8YFLogxK

U2 - 10.3233/IDA-150291

DO - 10.3233/IDA-150291

M3 - Article

AN - SCOPUS:85009905282

VL - 21

SP - 181

EP - 203

JO - Intelligent Data Analysis

JF - Intelligent Data Analysis

SN - 1088-467X

IS - 1

ER -