Prediction of hedonic tone using an electronic nose and artificial neural networks

Archie L. Williams, Paul H. Heinemann, Charles J. Wysocki, David M. Beyer, Robert E. Graves

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

An electronic nose was used in conjunction with a human panel and artificial neural networks to predict human assessments on a hedonic (pleasantness) scale of various odors, including fragrances, spent and phase I mushroom substrate odors. After calculating the average rating from the human panel for each odor sample, the hedonic rating was matched with the 32 electronic nose sensor readings from the corresponding sample. The data were then used to develop and train artificial neural networks. Several neural networks were developed and tested by comparing predicted hedonic tone from the neural networks with the average hedonic tone assessed by the human panel. The general regression neural network using 11 e-nose sensors as input gave the best test set results, yielding an r2 value of 0.92, and a root mean error of 0.06 hedonic tone units. The genetic strategy neural network gave the best validation results, with an r2 value of 0.67 and a root mean error of 1.11 pleasantness units. These results provide evidence that an electronic nose could potentially be used to assess the quality of odors in a similar manner to a human panel.

Original languageEnglish (US)
Pages (from-to)343-350
Number of pages8
JournalApplied Engineering in Agriculture
Volume26
Issue number2
StatePublished - Apr 28 2010

Fingerprint

Odors
Neural networks
Fragrances
Sensors
Electronic nose
Substrates

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Williams, Archie L. ; Heinemann, Paul H. ; Wysocki, Charles J. ; Beyer, David M. ; Graves, Robert E. / Prediction of hedonic tone using an electronic nose and artificial neural networks. In: Applied Engineering in Agriculture. 2010 ; Vol. 26, No. 2. pp. 343-350.
@article{4f9fe3a6d96646b1a74f8acbb5db7fc3,
title = "Prediction of hedonic tone using an electronic nose and artificial neural networks",
abstract = "An electronic nose was used in conjunction with a human panel and artificial neural networks to predict human assessments on a hedonic (pleasantness) scale of various odors, including fragrances, spent and phase I mushroom substrate odors. After calculating the average rating from the human panel for each odor sample, the hedonic rating was matched with the 32 electronic nose sensor readings from the corresponding sample. The data were then used to develop and train artificial neural networks. Several neural networks were developed and tested by comparing predicted hedonic tone from the neural networks with the average hedonic tone assessed by the human panel. The general regression neural network using 11 e-nose sensors as input gave the best test set results, yielding an r2 value of 0.92, and a root mean error of 0.06 hedonic tone units. The genetic strategy neural network gave the best validation results, with an r2 value of 0.67 and a root mean error of 1.11 pleasantness units. These results provide evidence that an electronic nose could potentially be used to assess the quality of odors in a similar manner to a human panel.",
author = "Williams, {Archie L.} and Heinemann, {Paul H.} and Wysocki, {Charles J.} and Beyer, {David M.} and Graves, {Robert E.}",
year = "2010",
month = "4",
day = "28",
language = "English (US)",
volume = "26",
pages = "343--350",
journal = "Applied Engineering in Agriculture",
issn = "0883-8542",
publisher = "American Society of Agricultural and Biological Engineers",
number = "2",

}

Prediction of hedonic tone using an electronic nose and artificial neural networks. / Williams, Archie L.; Heinemann, Paul H.; Wysocki, Charles J.; Beyer, David M.; Graves, Robert E.

In: Applied Engineering in Agriculture, Vol. 26, No. 2, 28.04.2010, p. 343-350.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Prediction of hedonic tone using an electronic nose and artificial neural networks

AU - Williams, Archie L.

AU - Heinemann, Paul H.

AU - Wysocki, Charles J.

AU - Beyer, David M.

AU - Graves, Robert E.

PY - 2010/4/28

Y1 - 2010/4/28

N2 - An electronic nose was used in conjunction with a human panel and artificial neural networks to predict human assessments on a hedonic (pleasantness) scale of various odors, including fragrances, spent and phase I mushroom substrate odors. After calculating the average rating from the human panel for each odor sample, the hedonic rating was matched with the 32 electronic nose sensor readings from the corresponding sample. The data were then used to develop and train artificial neural networks. Several neural networks were developed and tested by comparing predicted hedonic tone from the neural networks with the average hedonic tone assessed by the human panel. The general regression neural network using 11 e-nose sensors as input gave the best test set results, yielding an r2 value of 0.92, and a root mean error of 0.06 hedonic tone units. The genetic strategy neural network gave the best validation results, with an r2 value of 0.67 and a root mean error of 1.11 pleasantness units. These results provide evidence that an electronic nose could potentially be used to assess the quality of odors in a similar manner to a human panel.

AB - An electronic nose was used in conjunction with a human panel and artificial neural networks to predict human assessments on a hedonic (pleasantness) scale of various odors, including fragrances, spent and phase I mushroom substrate odors. After calculating the average rating from the human panel for each odor sample, the hedonic rating was matched with the 32 electronic nose sensor readings from the corresponding sample. The data were then used to develop and train artificial neural networks. Several neural networks were developed and tested by comparing predicted hedonic tone from the neural networks with the average hedonic tone assessed by the human panel. The general regression neural network using 11 e-nose sensors as input gave the best test set results, yielding an r2 value of 0.92, and a root mean error of 0.06 hedonic tone units. The genetic strategy neural network gave the best validation results, with an r2 value of 0.67 and a root mean error of 1.11 pleasantness units. These results provide evidence that an electronic nose could potentially be used to assess the quality of odors in a similar manner to a human panel.

UR - http://www.scopus.com/inward/record.url?scp=77951227157&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951227157&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77951227157

VL - 26

SP - 343

EP - 350

JO - Applied Engineering in Agriculture

JF - Applied Engineering in Agriculture

SN - 0883-8542

IS - 2

ER -