Classifier comparison and sensor selection for e-noses

M. Pardo, G. Sberveglieri, B. Sisk, N. Lewis

Research output: Contribution to conferencePaper

2 Citations (Scopus)

Abstract

Data derived from chemical vapor sensor arrays was analyzed using Fischer linear discriminant analysis (LDA) and multilayer perceptrons (MLP). Datasets produced by an e-nose based on polymer-carbon black sensors and those produced by a second e-nose based on five thin film metal oxide sensors was analyzed. The datasets were chosen to give classification problems of varying hardness: from the discrimination of two almost Gaussian distributed analytes to the discrimination of two analytes in the presents of interferents at different concentration levels. The performance of the best 5-sensors subset selected with MLP was found to be better than the performance of the LDA-selected subsets.

Original languageEnglish
Pages606-610
Number of pages5
Publication statusPublished - Dec 1 2003
EventSecond International Conference on Sensors: IEEE Sensors 2003 - Toronto, Ont., Canada
Duration: Oct 22 2003Oct 24 2003

Other

OtherSecond International Conference on Sensors: IEEE Sensors 2003
CountryCanada
CityToronto, Ont.
Period10/22/0310/24/03

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Classifier comparison and sensor selection for e-noses'. Together they form a unique fingerprint.

  • Cite this

    Pardo, M., Sberveglieri, G., Sisk, B., & Lewis, N. (2003). Classifier comparison and sensor selection for e-noses. 606-610. Paper presented at Second International Conference on Sensors: IEEE Sensors 2003, Toronto, Ont., Canada.