VOCs classification based on the committee of classifiers coupled with single sensor signals

Abstract A gas classification method based on a Multiple Classifiers System (MCS) is presented in this paper. The novelty of the approach consists in utilizing a signal of one sensor as the information source of a single member of the classifier ensemble. The size of the committee is delimited by the number of sensors applied for solving gas identification problems. The following base classifiers were considered: Support Vector Machine (SVM), the k -Nearest Neighbor ( k -NN) method and two kinds of decision trees — CART and C4.5. Additionally, three fusion strategies were examined: majority voting, weight assignment based on the individual accuracy of the committee member and optimal weights combination found by the genetic algorithm. The MCSs performance was compared with the effectiveness of single classifiers which operated on the data set containing the response of the entire sensor array. The sensor signal compression by means of granulation was applied as the data pre-processing step. The classification problem consisted in recognizing volatile organic compounds (VOCs) in air, based on measurements performed by the array composed of fifteen semiconductor gas sensors. These devices were operated in the stop flow mode. Thus their signals were affected by many factors associated with altering exposure conditions, which enhanced the discrimination abilities of the sensors.

[1]  Robert Burduk,et al.  Classification error in Bayes multistage recognition task with fuzzy observations , 2010, Pattern Analysis and Applications.

[2]  James C. Bezdek,et al.  Decision templates for multiple classifier fusion: an experimental comparison , 2001, Pattern Recognit..

[3]  Pietro Siciliano,et al.  Odor discrimination using adaptive resonance theory , 2000 .

[4]  Shankar Vembu,et al.  Chemical gas sensor drift compensation using classifier ensembles , 2012 .

[5]  Ingemar Lundström,et al.  Robust gas detection at sub ppm concentrations , 2011 .

[6]  J. Brezmes,et al.  Quantitative analysis of NO2 in the presence of CO using a single tungsten oxide semiconductor sensor and dynamic signal processing. , 2002, The Analyst.

[7]  Amine Bermak,et al.  Gas identification using density models , 2005, Pattern Recognit. Lett..

[8]  N. Ancona,et al.  Support vector machines for olfactory signals recognition , 2003 .

[9]  Amine Bermak,et al.  Pattern Recognition Techniques for Odor Discrimination in Gas Sensor Array , 2005 .

[10]  Michał Woźniak,et al.  Combining classifiers using trained fuser - analytical and experimental results / , 2010 .

[11]  Andrzej Szczurek,et al.  The stop-flow mode of operation applied to a single chemiresistor , 2010 .

[12]  Andreas Schütze,et al.  High performance solvent vapor identification with a two sensor array using temperature cycling and pattern classification , 2003 .

[13]  Kevin J. Johnson,et al.  A novel chemical detector using cermet sensors and pattern recognition methods for toxic industrial chemicals , 2006 .

[14]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[15]  A. Szczurek,et al.  Recognition of benzene, toluene and xylene using TGS array integrated with linear and non-linear classifier. , 2004, Talanta.

[16]  S. Phanichphant,et al.  Semiconducting metal oxides as sensors for environmentally hazardous gases , 2011 .

[17]  Antonella Macagnano,et al.  Electronic-nose modelling and data analysis using a self-organizing map , 1997 .

[18]  Francisco J. Ramirez-Fernandez,et al.  Committee machine for LPG calorific power classification , 2006 .

[19]  Eduard Llobet,et al.  Fuzzy ARTMAP based electronic nose data analysis , 1999 .

[20]  Nikolai F. Rulkov,et al.  Acceleration of chemo-sensory information processing using transient features , 2009 .

[21]  Véronique Bellon-Maurel,et al.  Optimisation of electronic nose measurements. Part I: Methodology of output feature selection , 1998 .

[22]  Rajeshuni Ramesham,et al.  Electronic nose for space program applications. , 2003, Sensors and actuators. B, Chemical.

[23]  Eduard Llobet,et al.  Response model for thermally modulated tin oxide-based microhotplate gas sensors , 2003 .

[24]  Ethem Alpaydın,et al.  Combined 5 x 2 cv F Test for Comparing Supervised Classification Learning Algorithms , 1999, Neural Comput..

[25]  S. Rose-Pehrsson,et al.  A comparison study of chemical sensor array pattern recognition algorithms , 1999 .

[26]  Pradeep Kurup,et al.  Decision tree approach for classification and dimensionality reduction of electronic nose data , 2011 .

[27]  C. Xie,et al.  An entire feature extraction method of metal oxide gas sensors , 2008 .

[28]  Russell Binions,et al.  Metal Oxide Semi-Conductor Gas Sensors in Environmental Monitoring , 2010, Sensors.

[29]  C. K. Chow,et al.  Statistical Independence and Threshold Functions , 1965, IEEE Trans. Electron. Comput..

[30]  D. E. Goldberg,et al.  Optimization and Machine Learning , 2022 .

[31]  A. Bermak,et al.  A Committee Machine Gas Identification System Based on Dynamically Reconfigurable FPGA , 2008, IEEE Sensors Journal.

[32]  D. Wolpert.  The Supervised Learning No-Free-Lunch Theorems , 2002 .

[33]  Michal Wozniak,et al.  Optimization of overlay distributed computing systems for multiple classifier system - heuristic approach , 2012, Log. J. IGPL.

[34]  R. Huerta,et al.  Multifrequency interrogation of nanostructured gas sensor arrays: a tool for analyzing response kinetics. , 2012, Analytical chemistry.

[35]  Shankar Vembu,et al.  On time series features and kernels for machine olfaction , 2012 .

[36]  Michal Wozniak,et al.  Complexity and Multithreaded Implementation Analysis of One Class-Classifiers Fuzzy Combiner , 2011, HAIS.

[37]  Bartosz Krawczyk,et al.  Designing Cost-Sensitive Ensemble - Genetic Approach , 2011, IP&C.

[38]  A Szczurek,et al.  Discrimination of coatings on wooden materials using the gas sensor system. , 2005, Talanta.

[39]  G. Sberveglieri,et al.  Comparing the performance of different features in sensor arrays , 2007 .

[40]  Thomas G. Dietterich,et al.  Solving Multiclass Learning Problems via Error-Correcting Output Codes , 1994, J. Artif. Intell. Res..

[41]  Weiguo Song,et al.  Detection of VOCs and their concentrations by a single SnO2 sensor using kinetic information , 2007 .

[42]  M. Pardo,et al.  Random forests and nearest shrunken centroids for the classification of sensor array data , 2008 .

[43]  Zulfiqur Ali,et al.  Data analysis for electronic nose systems , 2006 .

[44]  J. Goschnick,et al.  Gradient gas sensor microarrays for on-line process control — a new dynamic classification model for fast and reliable air quality assessment , 2000 .

[45]  Amine Bermak,et al.  Gas Identification Based on Committee Machine for Microelectronic Gas Sensor , 2006, IEEE Transactions on Instrumentation and Measurement.

[46]  C. Distante,et al.  On the study of feature extraction methods for an electronic nose , 2002 .

[47]  Stanislaw Osowski,et al.  Classification of gasoline with supplement of bio-products by means of an electronic nose and SVM neural network , 2006 .

[48]  B. Snopok,et al.  Multisensor systems for chemical analysis: state-of-the-art in Electronic Nose technology and new trends in machine olfaction , 2002 .

[49]  Kagan Tumer,et al.  Analysis of decision boundaries in linearly combined neural classifiers , 1996, Pattern Recognit..

[50]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[51]  Eduard Llobet,et al.  Selectivity Enhancement in Multisensor Systems Using Flow Modulation Techniques , 2008, Sensors.

[52]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[53]  A. Bermak,et al.  Fast and robust gas identification system using an integrated gas sensor technology and Gaussian mixture models , 2005, IEEE Sensors Journal.

[54]  D. E. Goldberg,et al.  Genetic Algorithms in Search , 1989 .

[55]  A. Hierlemann,et al.  Higher-order Chemical Sensing , 2007 .