CoChat
Sign Up
Cited by 2,739Machine Learning and Data ClassificationMachine Learning and AlgorithmsText and Document Classification Technologies

The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.

Solving Multiclass Learning Problems via Error-Correcting Output Codes

Open access
Dietterich, T. G., Bakiri, G.·1995·Journal of Artificial Intelligence Research
DOI 10.1613/jair.105
Open this paper in CoChat — free accountOpen PDF

Abstract

Multiclass learning problems involve finding a definitionfor an unknown function f(x) whose range is a discrete setcontaining k > 2 values (i.e., k ``classes''). Thedefinition is acquired by studying collections of training examples ofthe form [x_i, f (x_i)]. Existing approaches tomulticlass learning problems include direct application of multiclassalgorithms such as the decision-tree algorithms C4.5 and CART,application of binary concept learning algorithms to learn individualbinary functions for each of the k classes, and application ofbinary concept learning algorithms with distributed outputrepresentations. This paper compares these three approaches to a newtechnique in which error-correcting codes are employed as adistributed output representation. We show that these outputrepresentations improve the generalization performance of both C4.5and backpropagation on a wide range of multiclass learning tasks. Wealso demonstrate that this approach is robust with respect to changesin the size of the training sample, the assignment of distributedrepresentations to particular classes, and the application ofoverfitting avoidance techniques such as decision-tree pruning.Finally, we show that---like the other methods---the error-correctingcode technique can provide reliable class probability estimates.Taken together, these results demonstrate that error-correcting outputcodes provide a general-purpose method for improving the performanceof inductive learning programs on multiclass problems.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1613/jair.105

Cites

4 of 4 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • If the probabilistic assumptions in an HMM or other probabilistic model are not realistic, discriminative training can improve performance, as shown for speech recognition systems.

    “On the other hand, if the probabilistic assumptions in an HMM (or other probabilistic model) are not realistic, discriminative training, discussed in Section VI, can improve performance as this has been clearly shown for speech recognition systems [48]- [50], [107], [108].”

    From Gradient-based learning applied to document recognition (1998) · cites this paper as [48]

    The Introduction discusses phoneme recognition as a multiclass task, and Section 2.1 describes the NETtalk pronunciation dataset. Neither passage addresses unrealistic HMM assumptions or whether discriminative training improves speech-recognition performance; searches for those terms found no supporting passage in our copy. (LLM, conf=0.75)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 22

  1. Bridle, J. S. (1990). Training stochastic model recognition algorithms as networks can lead to maximum mutual information estimation of parameters. In Touretzky, D. S. (Ed.), Neural Information Processing Systems, Vol. 2, pp. 211{217 San Francisco, CA.
  2. Morgan Kaufmann.
  3. Cardie, C. (1993). Using decision trees to improve case-based learning. In Proceedings of the Tenth International Conference on Machine Learning, pp. 17{24 San Francisco, CA.
  4. Morgan Kaufmann.
  5. Freund, Y. (1992). An improved boosting algorithm and its implications on learning com- plexity. In Proc. 5th Annu. Workshop on Comput. Learning Theory, pp. 391{398. ACM Press, New York, NY.
  6. Hastie, T., Tibshirani, R., & Buja, A. (In Press). Flexible discriminant analysis by optimal scoring. Journal of the American Statistical Association.
  7. Lang, K. J., Hinton, G. E., & Waibel, A. (1990). A time-delay neural network architecture for isolated word recognition. Neural Networks, 3(1), 23{43.
  8. Natarajan, B. K. (1991). Machine Learning: A Theoretical Approach. Morgan Kaufmann, San Mateo, CA.
  9. Nilsson, N. J. (1965). Learning Machines. McGraw-Hill, New York.
  10. Perrone, M. P., & Cooper, L. N. (1993). When networks disagree: Ensemble methods for hybrid neural networks. In Mammone, R. J. (Ed.), Neural networks for speech and image processing. Chapman and Hall.
  11. Peterson, W. W., & Weldon, Jr., E. J. (1972). Error-Correcting Codes. MIT Press, Cam- bridge, MA.
  12. Morgan Kaufmann, San Francisco, CA.
  13. Richard, M. D., & Lippmann, R. P. (1991). Neural network classi ers estimate bayesian a posteriori probabilities. Neural Computation, 3(4), 461{483.
  14. Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning internal representa- tions by error propagation. In Parallel Distributed Processing { Explorations in the Microstructure of Cognition, chap. 8, pp. 318{362. MIT Press.
  15. Sejnowski, T. J., & Rosenberg, C. R. (1987). Parallel networks that learn to pronounce english text. Journal of Complex Systems, 1(1), 145{168.
  16. Selman, B., Levesque, H., & Mitchell, D. (1992). A new method for solving hard satis ability problems. In Proceedings of AAAI-92, pp. 440{446. AAAI/MIT Press.
  17. Snedecor, G. W., & Cochran, W. G. (1989). Statistical Methods. Iowa State University Press, Ames, IA. Eighth Edition.
  18. Valiant, L. G. (1984). A theory of the learnable. Commun. ACM, 27(11), 1134{1142.
  19. Waibel, A., Hanazawa, T., Hinton, G., Shikano, K., & Lang, K. (1989). Phoneme recogni- tion using time-delay networks. IEEE Transactions on Acoustics, Speech, and Signal Processing, 37(3), 328{339.
  20. Weigend, A. (1993). Measuring the eective number of dimensions during backpropagation training. In Proceedings of the 1993 Connectionist Models Summer School, pp. 335{ 342.
  21. Morgan Kaufmann, San Francisco, CA.
  22. Wilkinson, R. A., Geist, J., Janet, S., et al. (1992). The rst census optical character recog- nition systems conference. Tech. rep. NISTIR 4912, National Institute of Standards and Technology.

Cited by

Sorted by citations

Metadata

Authors
Dietterich, T. G., & Bakiri, G.
Affiliations
University of Bahrain (BH)
ORCID
Dietterich, T. G.
Venue
Journal of Artificial Intelligence Research
Publisher
AI Access Foundation
Published
1995-01-01
Volume
2
Pages
263-286
ISSN
1076-9757 (ISSN-L)
DOI
10.1613/jair.105
OpenAlex
W1676820704
Cited by
2,739
References
4
Citations by year
2026: 39 · 2025: 45 · 2024: 50 · 2023: 64 · 2022: 74 · 2021: 104 · 2020: 128 · 2019: 110 · 2018: 121 · 2017: 116 · 2016: 99 · 2015: 127 · 2014: 146 · 2013: 129 · 2012: 134
FWCI
18.29
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Journal of Artificial Intelligence Research (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Funders
National Science Foundation (9216172, 9204129)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
  • Physical Sciences › Computer Science › Artificial Intelligence › Text and Document Classification Technologies
Keywords
error-correcting output codes, C4.5, CART, backpropagation, generalization performance, decision tree pruning, overfitting avoidance, class probability estimation, inductive learning

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review