Generation of Sets of Synthetic Classifiers for the Evaluation of Abstract-Level Combination Methods
This paper presents a new technique for generating sets of synthetic classifiers to evaluate abstract-level combination methods. The sets differ in terms of both recognition rates of the individual classifiers and degree of similarity. For this purpose, each abstract-level classifier is considered as a random variable producing one class label as the output for an input pattern. From the initial set of classifiers, new slightly different sets are generated by applying specific operators, which are defined at the purpose. Finally, the sets of synthetic classifiers have been used to estimate the performance of combination methods for abstract-level classifiers. The experimental results demonstrate the effectiveness of the proposed approach.
[1] L. Xu, A. Krzyzak, C. Y-Suen, "Methods of Combining Multiple
Classifiers and Their Applications to Handwriting Recognition",IEEE TSMC,
Vol.22, N.3,1992, pp. 418-435.
[2] J. Kittler, M. Hatef, R.P.W. Duin and J. Matias, "On combining
classifiers", IEEE T-PAMI, vol. 20, no. 3, pp. 226-239, 1998.
[3] E.M.Kleinberg,"Stochastic Discimination",Annals of Math. ematics and
Artificial Intelligence, Vol. 1, pp. 207-239, 1990.
[4] S.N. Srihari, "Reliability Analysis of Majority Vote Systems",
Information Sciences, Vol. 26, pp. 243-256, 1982.
[5] L.Lam, "Classifier Cotions: Implementations and Theoretical Issues", in
Multiple Classifier Systems, J. Kittler an F. Roli (eds.), LNCS, vol. 1857,
Springer, 2000, pp. 77-86.
[6] K. Tumer and J. Ghosh, "Analysis of Decision Boundaries in inearly
Combined Neural Classifiers", Pattern Recognition, Vol. 29, pp. 341-348,
1996.
[7] H. Zouari, L. Heutte, Y. Lecourtier, A. Alimi, " A New Classifier
Simulator for Evaluating Parallel Combination Methods", Proc. ICDAR
2003, Edinburgh, UK, 2003, pp.26-30.
[8] J.R. Parker,"Rank and response combination from confusion matrix
data", Information Fusion, Vol. 2, 2001, pp. 113-120.
[9] L.I.Kuncheva and R.K. Kuntchev, "Generating classifier outputs of
fixed accuracy and diversity", PRL, Vol. 23, 2002, pp. 593-600.
[10] L. Bovino, G. Dimauro, S.Impedovo, M.G. Lucchese, R. Modugno, G.
Pirlo A.Salzo, L. Sarcinella, "On the Combination of Abstract-Level
Classifiers", IJDAR, 2003, Vol. 6, pp. 42-54.
[11]G. Dimauro, S. Impedovo, G. Pirlo, A. Salzo, "Multiple Classifiers: a
new methodology for the evaluation of the combination processes",
Progress in Handwriting Recognition, A.C.Downton and S. Impedovo
(eds.), World Scient.,Singapore, pp.329-335, 1995.
[12]V. Di Lecce,G.Dimauro,A.Guerriero,S.Impedovo,G.Pirlo,A. Salzo,
"Knowledge-based Methods for Classifier Combination: an Experimental
Investigation", Proc.ICIAP-7,Italy,1999, pp.562-565.
[13]S. Impedovo, G. Pirlo, "The Similarity Index", Università degli Studi di
Bari, Technical Report TC. 1, 2002.
[14]E. Mandler and J. Schuermann, "Combining the Classification Results of
independent classifiers based on the Dempster/Shafer theory of
evidence", in Pattern Recognition and Artificial Intelligence, North
Holland, Amsterdam, 1988, pp. 381-393.
[15]G. Dimauro, S. Impedovo, G. Pirlo, A. Salzo, "Bankcheck recognition
systems: re-engineering the design process". In Progress in Handwriting
Recognition, A.C. Downton and S. Impedovo (eds.), World Scientific
Publ., 1997, pp. 419-425.
[1] L. Xu, A. Krzyzak, C. Y-Suen, "Methods of Combining Multiple
Classifiers and Their Applications to Handwriting Recognition",IEEE TSMC,
Vol.22, N.3,1992, pp. 418-435.
[2] J. Kittler, M. Hatef, R.P.W. Duin and J. Matias, "On combining
classifiers", IEEE T-PAMI, vol. 20, no. 3, pp. 226-239, 1998.
[3] E.M.Kleinberg,"Stochastic Discimination",Annals of Math. ematics and
Artificial Intelligence, Vol. 1, pp. 207-239, 1990.
[4] S.N. Srihari, "Reliability Analysis of Majority Vote Systems",
Information Sciences, Vol. 26, pp. 243-256, 1982.
[5] L.Lam, "Classifier Cotions: Implementations and Theoretical Issues", in
Multiple Classifier Systems, J. Kittler an F. Roli (eds.), LNCS, vol. 1857,
Springer, 2000, pp. 77-86.
[6] K. Tumer and J. Ghosh, "Analysis of Decision Boundaries in inearly
Combined Neural Classifiers", Pattern Recognition, Vol. 29, pp. 341-348,
1996.
[7] H. Zouari, L. Heutte, Y. Lecourtier, A. Alimi, " A New Classifier
Simulator for Evaluating Parallel Combination Methods", Proc. ICDAR
2003, Edinburgh, UK, 2003, pp.26-30.
[8] J.R. Parker,"Rank and response combination from confusion matrix
data", Information Fusion, Vol. 2, 2001, pp. 113-120.
[9] L.I.Kuncheva and R.K. Kuntchev, "Generating classifier outputs of
fixed accuracy and diversity", PRL, Vol. 23, 2002, pp. 593-600.
[10] L. Bovino, G. Dimauro, S.Impedovo, M.G. Lucchese, R. Modugno, G.
Pirlo A.Salzo, L. Sarcinella, "On the Combination of Abstract-Level
Classifiers", IJDAR, 2003, Vol. 6, pp. 42-54.
[11]G. Dimauro, S. Impedovo, G. Pirlo, A. Salzo, "Multiple Classifiers: a
new methodology for the evaluation of the combination processes",
Progress in Handwriting Recognition, A.C.Downton and S. Impedovo
(eds.), World Scient.,Singapore, pp.329-335, 1995.
[12]V. Di Lecce,G.Dimauro,A.Guerriero,S.Impedovo,G.Pirlo,A. Salzo,
"Knowledge-based Methods for Classifier Combination: an Experimental
Investigation", Proc.ICIAP-7,Italy,1999, pp.562-565.
[13]S. Impedovo, G. Pirlo, "The Similarity Index", Università degli Studi di
Bari, Technical Report TC. 1, 2002.
[14]E. Mandler and J. Schuermann, "Combining the Classification Results of
independent classifiers based on the Dempster/Shafer theory of
evidence", in Pattern Recognition and Artificial Intelligence, North
Holland, Amsterdam, 1988, pp. 381-393.
[15]G. Dimauro, S. Impedovo, G. Pirlo, A. Salzo, "Bankcheck recognition
systems: re-engineering the design process". In Progress in Handwriting
Recognition, A.C. Downton and S. Impedovo (eds.), World Scientific
Publ., 1997, pp. 419-425.
@article{"International Journal of Information, Control and Computer Sciences:60259", author = "N. Greco and S. Impedovo and R.Modugno and G. Pirlo", title = "Generation of Sets of Synthetic Classifiers for the Evaluation of Abstract-Level Combination Methods", abstract = "This paper presents a new technique for generating sets of synthetic classifiers to evaluate abstract-level combination methods. The sets differ in terms of both recognition rates of the individual classifiers and degree of similarity. For this purpose, each abstract-level classifier is considered as a random variable producing one class label as the output for an input pattern. From the initial set of classifiers, new slightly different sets are generated by applying specific operators, which are defined at the purpose. Finally, the sets of synthetic classifiers have been used to estimate the performance of combination methods for abstract-level classifiers. The experimental results demonstrate the effectiveness of the proposed approach.
", keywords = "Abstract-level Classifier, Dempster-Shafer Rule, Multi-expert Systems, Similarity Index, System Evaluation", volume = "1", number = "6", pages = "1767-4", }
