This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Classification in Noisy Environments Using a Distance Measure Between Structural Symbolic Descriptions
March 1992 (vol. 14 no. 3)
pp. 390-402

A definition of distance measure between structural descriptions that is based on a probabilistic interpretation of the matching predicate is proposed. It aims at coping with the problem of classification when noise causes both local and structural deformations. The distance measure is defined according to a top-down evaluation scheme: distance between disjunctions of conjuncts, conjunctions, and literals. At the lowest level, the similarity between a feature value in the pattern model (G) and the corresponding value in the observation (Ex) is defined as the probability of observing a greater distortion. The classification problem is approached by means of a multilayered framework in which the cases of single perfect match, no perfect match, and multiple perfect match are treated differently. A plausible solution for the problem of completing the attribute and structure spaces, based on the probabilistic approach, is also given. A comparison with other related works and an application in the domain of layout-based document recognition are presented.

[1] B. Hayes-Roth, "Implications of human pattern processing for the design of artificial knowledge systems," inPattern-Directed Inference Systems(D.A. Waterman and F. Hayes-Roth, Eds.). Orlando: Academic, 1978, pp. 333-346.
[2] D. H. Ballard and C. M. Brown,Computer Vision. Englewood Cliffs, NJ: Prentice-Hall, 1982.
[3] A. Sanfeliu and K. S. Fu, "A distance measure between attributed relational graphs for pattern recognition,"IEEE Trans. Syst. Man Cybern., vol. SMC-13, pp. 353-362, 1983.
[4] A. K. C. Wong and M. You, "Entropy and distance of random graphs with application to structural pattern recognition,"IEEE Trans. Patt. Anal. Machine Intell., vol. PAMI-7, pp. 599-609, 1985.
[5] J. R. Quinlan, "The effect of noise on concept learning," inMachine Learning: An Artificial Intelligence Approach, Vol. II(R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, Eds.). Los Altos: Morgan Kaufmann, 1986.
[6] R. S. Michalski, I. Mozetic, J. Hong, and N. Lavrac, "The AQ15 inductive learning system: An overview and experiments," Intell. Syst. Group, Dept. of Comput. Sci., Univ. of Illinois, Urbana, IL, 1986.
[7] F. Bergadano, A. Giordana, and L. Saitta, "Automated concept acquisition in noisy environments,"IEEE Trans. Patt. Anal. Machine Intell., vol. 10, pp. 555-578, 1988.
[8] D. J. Hand,Discrimination and Classification: London, Wiley, 1981.
[9] R. S. Michalski and J. B. Larson, "Selection of the most representative training examples and incremental generation ofV L1hypotheses: The underlying methodology and the description of programs ESEL and AQ11," Tech. Rep. UIUCDCS-R-78-867, Dept. of Comput. Sci., Univ. of Illinois, Urbana, IL, 1978.
[10] W. H. Tsai and K. S. Fu, "Subgraph error-correcting isomorphisms for syntactic pattern recognition,"IEEE Trans. Syst. Man Cybern., vol. SMC-13, pp. 48-62, 1983.
[11] M. A. Eshera and K. S. Fu, "A graph distance measure for image analysis,"IEEE Trans. Syst. Man Cybern., vol. SMC-14, pp. 398-408, 1984.
[12] Y. Kodratoff and G. Tecuci, "Learning based on conceptual distance,"IEEE Trans. Patt. Anal. Machine Intell., vol. 10, pp. 897-909, 1988.
[13] R. E. Stepp, "Machine learning from structured objects," inProc. Fourth Int. Workshop Machine Learning. Irvine, CA: Morgan Kaufman, 1987.
[14] R. S. Michalski, "Pattern recognition as rule-guided inductive inference,"IEEE Trans. Patt. Anal. Machine Intell., vol. PAMI-2, pp. 349-361, 1980.
[15] R. E. Stepp, "Conjunctive conceptual clustering: A methodology and experimetation," Doctoral dissertation, Rep. UIUCDCS-R-84-1189, Dept. of Comput. Sci., Univ. of Illinois, Urbana, IL, 1984.
[16] J. B. Larson, "Inductive inference in the variable valued predicate logic systemV L21: Methodology and computer implementation," Doctoral dissertation, Dept. of Comput. Sci., Univ. of Illinois, Urbana, IL, 1977.
[17] D. G. Bobrow, "Dimensions of representation," inRepresentation and Understanding(D. G. Bobrow and A. Collins, Eds.). New York: Academic, 1975, pp. 1-34.
[18] M. G. Thomason, "Structural methods in pattern analysis," inPattern Recognition Theory and Applications(P.A. Devijver and J. Kittler, Eds.). Berlin: Springer-Verlag, 1987.
[19] M. R. Garey and D. S. Johnson,Computers and Intractability. San Francisco, CA: W.H. Freeman, 1979.
[20] R. S. Michalski and R. E. Stepp, "Revealing conceptual structure in data by inductive inference," inMachine Intelligence. (J. E. Hayes, D. Michie, Y. -H. Pao, Eds.). London: Ellis Horwood, 1982.
[21] P. M. Winston,Artificial Intelligence. Reading, MA: Addison-Wesley, 1984.
[22] L. G. Shapiro and R. M. Haralick, "A metric for comparing relational descriptions,"IEEE Trans. Patt. Anal. Machine Intell., vol. PAMI-2, pp. 90-94, 1985.
[23] V. Barnett,Comparative Statistical Inference. London: Wiley, 1973.
[24] L. S. Chan and O. J. Dunn, "The treatment of missing values in discriminant analysis-I. The sampling experiment,"J. Amer. Stat. Assoc., vol. 67, pp. 473-477, 1972.
[25] F. Esposito, D. Malerba, G. Semeraro, E. Annese, and G. Scafuro, "An experimental page layout recognition system for office document automatic classification: An integrated approach for inductive generalization," inProc. 10th IEEE Int. Conf. Patt. Recogn.(Atlantic City, NJ), 1990, pp. 557-562.
[26] R. S. Michalski, "On the quasi-minimal solution of the general covering problem," inProc. V Int. Symp. Inform. Processing (FCIP 69)(Bled, Yugoslavia), 1969, vol. A3 (Switching Circuits).
[27] R. S. Michalski, "V L1: variable-valued logic system," inProc. 1974 Int. Symp. Multiple-Valued Logic(Morgantown, WV), 1974.
[28] R. S. Michalski, "How to learn imprecise concepts: A method for employing a two-tiered knowledge representation in learning," inProc. Fourth Int. Workshop Machine Learning. Irvine, CA: Morgan Kaufman, 1987.

Index Terms:
attribute spaces; pattern recognition; learning systems; noisy environments; distance measure; structural symbolic descriptions; probabilistic interpretation; matching predicate; top-down evaluation; feature value; pattern model; classification; structure spaces; layout-based document recognition; learning systems; pattern recognition; probability
Citation:
F. Esposito, D. Malerba, G. Semeraro, "Classification in Noisy Environments Using a Distance Measure Between Structural Symbolic Descriptions," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 3, pp. 390-402, March 1992, doi:10.1109/34.120333
Usage of this product signifies your acceptance of the Terms of Use.