Special Issue on Neuro-Computing and Hybrid Methods for Evolving IntelligenceNo Access

EVALUATION OF SELECTING INTERVAL VALUES OF INPUT VARIABLES IN CONNECTIONIST NETWORKS

DAVID K. Y. CHIU

Department of Computing & Information Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1, Canada

Search for more papers by this author

and

BOGDAN J. BUCZYNSKI

Department of Computing & Information Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1, Canada

Search for more papers by this author

https://doi.org/10.1142/S146902680400129XCited by:0 (Source: Crossref)

Abstract

Selecting parameters can be a powerful mechanism in constructing new evolving connectionist network. However, if a parameter contains partial information such that only some of the values are relevant and others are not, then a selection of the subset of relevant values is more appropriate. Considering the possible values of a parameter of a processing connectionist network as the outcomes of a variable, this research focuses on selecting interval values of the variable. It also considers the partitioning schemes used in generating the intervals from the outcomes of a variable. The goal of this work is to explore variable value selection and its effect in an evolving connectionist network. Using input variables in a backpropagation network, the proposed method evaluates its effect based on training of a dataset, and eliminates those intervals of the variable values that contribute negatively when processed by the network. When a value falls into an interval that has been selected and ignored, it is analogous to a network without processing the corresponding variable, and vice versa. Two approaches for interval partitioning are considered, based on equal-probability (or maximum entropy) and equal-width partitioning scheme. Comparing the best performing network with selection and the one without selection, the experimental results show that the best network with selection can produce better performance accuracy and smaller network size.

Keywords:

Remember to check out the Most Cited Articles!
Check out these titles in artificial intelligence!

References

H. Liu , Feature Selection for Knowledge Discovery and Data Mining ( Kluwer Academic Publishers , Boston, MA , 1998 ) . Crossref, Google Scholar
D. Aha and R. L. Bankert, A comparative evaluation of sequential feature selection algorithms, Proc. 5th Int. Workshop Artif. Intel. Statist. (1995) pp. 1–7. Google Scholar
D. Koller and M. Sahami, Toward optimal feature selection, Proc. 13th Int. Conf. Much. Learning (1996) pp. 284–292. Google Scholar
G. H. John, R. Kohavi and K. Pfleger, Irrelevant features and the subset selection problem, Proc. 11th Int. Conf. Mach. Learning (1994) pp. 121–129. Google Scholar
D. K. Y. Chiu and M. Y. W. Leung, Evaluating input dependency in feedforward neural network, Proc. 5th Int. Conf. Neural Inf. Process. (1998) pp. 801–804. Google Scholar
Y. Kim, W. Street and F. Menczer, Feature selection in unsupervised learning via evolutionary search, Proc. 6th ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining (2000) pp. 365–369. Google Scholar
M. A. Hall, Correlation-based feature selection for discrete and numeric class machine learning, Proceedings 17th International Conference on Machine Learning (Morgan Kaufmann, 2000) pp. 359–366. Google Scholar
K. Messer and J. Kittler, Pattern Rec. Lett. 18, 1241 (1997), DOI: 10.1016/S0167-8655(97)00089-5. Crossref, Google Scholar
S. Davies and S. Russell, NP-completeness of searches for smallest possible feature sets, Proceedings of 1994 AAAI Fall Symposium on Relevance (AAAI Press, 1994) pp. 37–39. Google Scholar
A. K. C. Wong and D. K. Y. Chiu, IEEE Trans. Pattern Anal. Much. Intell. 9, 796 (1987). Google Scholar
D. K. Y. Chiu, B. Cheung and A. K. C. Wong, J. Exp. Theor. Art. Intell. 2, 117 (1990), DOI: 10.1080/09528139008953718. Crossref, Google Scholar
C. Bie, H. Shen and D. K. Y. Chiu, Pattern Recog. Lett. 14, 421 (1993), DOI: 10.1016/0167-8655(93)90121-S. Crossref, Google Scholar
H. Liu and R. Setiono, IEEE Trans. Knowledge Data Eng. 9, 642 (1997). Google Scholar
D. J. C. MacKay , Information Theory, Inference, and Learning Algorithms ( Cambridge University Press , 1999 ) . Google Scholar
J. W. Smithet al., Using the ADAP learning algorithm to forecast the onset of diabetes mellitus, Proceedings of Symposium Computation Applications and Medical Care (IEEE Computer Society Press, 1988) pp. 261–265. Google Scholar
P. L. Rosin and F. Fierens, Improving neural network generalization, Proc. Geosci. Remote Sensing Symp. IGARSS'95 (1995) pp. 1255–1257. Google Scholar
I. Sekitaet al., Adaptive and Learning Systems II 1962, ed. F. A. Sadjadi (1993) pp. 45–53. Crossref, Google Scholar