Abstract
We provide a model of concept drift by which concept classes are automatically updated according to the dynamics of change in the underlying learning environment. Instances with a well-defined degree of feature similarity are incrementally grouped together to form so-called concept versions. A comparative empirical evaluation of the versioning system VerGene, which is built on these premises, demonstrates its effectiveness for properly dealing with concept drift phenomena in medium-sized knowledge bases.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
D. W. Aha and R. L. Goldstone. Concept learning and flexible weighting. In Proc. of the 14th Annual Conf of the Cognitive Science Society, pages 534–539. L. Erlbaum,1992.
D. W. Aha, D. Kibler, and M. Albert. Instance-based learning algorithms. Machine Learning, 6:37–66, 1991.
R. H. Creecy, B. M. Mansand, S. J. Smith, and D. L. Waltz. Trading MIPS and memory for knowledge engineering. Communications of the ACM, 35:48–64, 1992.
W. Daelemans and A. van den Bosch. Generalization performance of backpropagation learning on a syllabification task. In Proc. of the TWLT 3 — Connectionism and Natural Language Processing, pages 27–37,1992.
K. Fukunaga and T. Flick. An optimal global nearest neighbor metric. IEEE Transactions on Pattern Analysis and Machine Intelligence, 6:314–318, 1984.
J. H. Gennari, P. Langley, and D. Fisher. Models of incremental concept formation. Artificial Intelligence, 40(1/3):11–61, 1989.
U. Hahn, S. Schacht, and N. Bröker. Concurrent, object-oriented natural language parsing: the Parsetalk model. Int'l. Journal of Human-Computer Studies, 41(1/2):179–222, 1994.
U. Hahn and K. Schnattinger. Deep knowledge discovery from natural language texts. In KDD '97 — Proc. of the 3rd Int'l. Conf. on Knowledge Discovery and Data Mining, 1997.
D. P. Helmbold and P. M. Long. Tracking drifting concepts by minimizing disagreements. Machine Learning, 14(1):27–45, 1994.
J. D. Kelly and L. Davis. A hybrid genetic algorithm for classification. In IJCAI '91 — Proc. of the 12th Int'l. Joint Conf. on Artificial Intelligence, pages 645–650,1991.
D. Kibler, D. W. Aha, and M. Albert. Instance-based prediction of real-valued attributes. Computational Intelligence, 5:51–57, 1989.
F. Kilander and C. G. Jansson. COBBIT: a control procedure for COBWEB in the presence of concept drift. In ECML '93 — Proc. of the European Conf. on Machine Learning, pages 245–261,1993.
M. Klenner and U. Hahn. Concept versioning: a methodology for tracking evolutionary concept drift in dynamic concept systems. In ECAI '94 — Proc. of the 11th European Conference on Artificial Intelligence, pages 473–477. J. Wiley, 1994.
T. Kuh, T Petsche, and R. Rivest. Incrementally learning time-varying half-planes. Neural Information Processing Systems, 4, 1992.
P. Langley. A general theory of discrimination learning. In D. Klahr, P Langley, and R. Neches, editors, Production System Models of Learning and Development, pages 99–161. MIT Press, 1987.
P M. Murphy and D. W. Aha. Automobile database from the UCI repository. http://www. ics.uci.edu/mlearn/MLRepository.html, 1994.
J. C. Schlimmer and R. H. Granger. Incremental learning from noisy data. Machine Learning, 1(3):317–354, 1986.
D. Skalak. Protoype and feature selection by sampling and random mutation hill climbing algorithms. In Proc. of the 11th Int'l. Machine Learning Conference, pages 293–301,1994.
J. Stutz and P Cheeseman. AutoClass: a Bayesian approach to classification. In J. Skilling and S. Sibisi, editors, Maximum Entropy and Bayesian Methods. Kluwer, 1994.
D. Wettscherek, D. W. Aha, and T. Mohri. A review and comparative evaluation of feature weighting methods for lazy learning algorithms. Technical Report AIC-95-012, Washington, DC: Naval Research Laboratory, Navy Center for Appl. Res. in Artificial Intelligence, 1995.
G. Widmer. Combining robustness and flexibility in learning drifting concepts. In ECAI '94 — Proc. of the 11th European Conference on Artificial Intelligence, pages 468–472, 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hahn, U., Klenner, M. (1997). Incremental concept evolution based on adaptive feature weighting. In: Coasta, E., Cardoso, A. (eds) Progress in Artificial Intelligence. EPIA 1997. Lecture Notes in Computer Science, vol 1323. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0023910
Download citation
DOI: https://doi.org/10.1007/BFb0023910
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63586-4
Online ISBN: 978-3-540-69605-6
eBook Packages: Springer Book Archive