Committee machine with over 95% classification accuracy for combustible gas identification

  • Shi Minghua*
  • , Amine Bermak
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

Gas identification represents a big challenge for pattern recognition systems due to several particular problems such as non-selectivity and drift. This paper proposes a gas identification committee machine (CM), which combines various gas identification algorithms, to obtain a unified decision with improved accuracy. The CM combines 5 different classifiers: K Nearest Neighbors (KNN), Multi-layer Perceptron (MLP), Radial Basis Function (RFB), Gaussian Mixture Model (GMM) and Probabilistic PCA (PPCA). A data acquisition system using tin-oxide gas sensor array has been designed in order to create a real gas data set. The committee machine is implemented by assembling the outputs of these gas identification algorithms based on weighted combination rule. Experiments on real sensors' data proved the effectiveness of our system with an improved accuracy 95.9% over the individual classifiers.

Original languageEnglish
Title of host publicationICECS 2006 - 13th IEEE International Conference on Electronics, Circuits and Systems
Pages862-865
Number of pages4
DOIs
Publication statusPublished - 2006
Externally publishedYes
EventICECS 2006 - 13th IEEE International Conference on Electronics, Circuits and Systems - Nice, France
Duration: 10 Dec 200613 Dec 2006

Publication series

NameProceedings of the IEEE International Conference on Electronics, Circuits, and Systems

Conference

ConferenceICECS 2006 - 13th IEEE International Conference on Electronics, Circuits and Systems
Country/TerritoryFrance
CityNice
Period10/12/0613/12/06

Keywords

  • Committee machine
  • Gas identification
  • Pattern recognition
  • Tin oxide gas sensor

Fingerprint

Dive into the research topics of 'Committee machine with over 95% classification accuracy for combustible gas identification'. Together they form a unique fingerprint.

Cite this