Getting the Most from Flawed Theories

Moshe Koppel; Alberto Maria Segre; Ronen Feldman

doi:10.1016/b978-1-55860-335-6.50025-8

Getting the Most from Flawed Theories

Moshe Koppel, Alberto Maria Segre, Ronen Feldman

Cornell University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

This paper introduces a new classification technique called degree-of-provedness classification, or DOP-classification. This technique exploits information implicit in the structure of a possibly incomplete or incorrect domain theory in order to improve classification accuracy. It is also shown how DOP-classification can be used to identify theories for which theory revision is unnecessary (because the unrevised theory can be used directly by DOP-classification to achieve near-perfect classification accuracy) or insufficient (because the initial theory is so flawed that it would be preferable to induce a new theory directly from examples).

Original language	English
Title of host publication	Proceedings of the 11th International Conference on Machine Learning, ICML 1994
Editors	William W. Cohen, Haym Hirsh
Publisher	Morgan Kaufmann Publishers, Inc.
Pages	139-147
Number of pages	9
ISBN (Electronic)	1558603352, 9781558603356
DOIs	https://doi.org/10.1016/b978-1-55860-335-6.50025-8
State	Published - 1994
Event	11th International Conference on Machine Learning, ICML 1994 - New Brunswick, United States Duration: 10 Jul 1994 → 13 Jul 1994

Publication series

Name	Proceedings of the 11th International Conference on Machine Learning, ICML 1994

Conference

Conference	11th International Conference on Machine Learning, ICML 1994
Country/Territory	United States
City	New Brunswick
Period	10/07/94 → 13/07/94

Bibliographical note

Publisher Copyright:
© 1994 Proceedings of the 11th International Conference on Machine Learning, ICML 1994. All rights reserved.

Access to Document

10.1016/b978-1-55860-335-6.50025-8

Cite this

Koppel, M., Segre, A. M., & Feldman, R. (1994). Getting the Most from Flawed Theories. In W. W. Cohen, & H. Hirsh (Eds.), Proceedings of the 11th International Conference on Machine Learning, ICML 1994 (pp. 139-147). (Proceedings of the 11th International Conference on Machine Learning, ICML 1994). Morgan Kaufmann Publishers, Inc.. https://doi.org/10.1016/b978-1-55860-335-6.50025-8

@inproceedings{9aa4f347eeff4c43b35d0288bf2d762c,

title = "Getting the Most from Flawed Theories",

abstract = "This paper introduces a new classification technique called degree-of-provedness classification, or DOP-classification. This technique exploits information implicit in the structure of a possibly incomplete or incorrect domain theory in order to improve classification accuracy. It is also shown how DOP-classification can be used to identify theories for which theory revision is unnecessary (because the unrevised theory can be used directly by DOP-classification to achieve near-perfect classification accuracy) or insufficient (because the initial theory is so flawed that it would be preferable to induce a new theory directly from examples).",

author = "Moshe Koppel and Segre, \{Alberto Maria\} and Ronen Feldman",

note = "Publisher Copyright: {\textcopyright} 1994 Proceedings of the 11th International Conference on Machine Learning, ICML 1994. All rights reserved.; 11th International Conference on Machine Learning, ICML 1994 ; Conference date: 10-07-1994 Through 13-07-1994",

year = "1994",

doi = "10.1016/b978-1-55860-335-6.50025-8",

language = "אנגלית",

series = "Proceedings of the 11th International Conference on Machine Learning, ICML 1994",

publisher = "Morgan Kaufmann Publishers, Inc.",

pages = "139--147",

editor = "Cohen, \{William W.\} and Haym Hirsh",

booktitle = "Proceedings of the 11th International Conference on Machine Learning, ICML 1994",

address = "ארצות הברית",

}

Koppel, M, Segre, AM & Feldman, R 1994, Getting the Most from Flawed Theories. in WW Cohen & H Hirsh (eds), Proceedings of the 11th International Conference on Machine Learning, ICML 1994. Proceedings of the 11th International Conference on Machine Learning, ICML 1994, Morgan Kaufmann Publishers, Inc., pp. 139-147, 11th International Conference on Machine Learning, ICML 1994, New Brunswick, United States, 10/07/94. https://doi.org/10.1016/b978-1-55860-335-6.50025-8

Getting the Most from Flawed Theories. / Koppel, Moshe; Segre, Alberto Maria; Feldman, Ronen.
Proceedings of the 11th International Conference on Machine Learning, ICML 1994. ed. / William W. Cohen; Haym Hirsh. Morgan Kaufmann Publishers, Inc., 1994. p. 139-147 (Proceedings of the 11th International Conference on Machine Learning, ICML 1994).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Getting the Most from Flawed Theories

AU - Koppel, Moshe

AU - Segre, Alberto Maria

AU - Feldman, Ronen

PY - 1994

Y1 - 1994

N2 - This paper introduces a new classification technique called degree-of-provedness classification, or DOP-classification. This technique exploits information implicit in the structure of a possibly incomplete or incorrect domain theory in order to improve classification accuracy. It is also shown how DOP-classification can be used to identify theories for which theory revision is unnecessary (because the unrevised theory can be used directly by DOP-classification to achieve near-perfect classification accuracy) or insufficient (because the initial theory is so flawed that it would be preferable to induce a new theory directly from examples).

AB - This paper introduces a new classification technique called degree-of-provedness classification, or DOP-classification. This technique exploits information implicit in the structure of a possibly incomplete or incorrect domain theory in order to improve classification accuracy. It is also shown how DOP-classification can be used to identify theories for which theory revision is unnecessary (because the unrevised theory can be used directly by DOP-classification to achieve near-perfect classification accuracy) or insufficient (because the initial theory is so flawed that it would be preferable to induce a new theory directly from examples).

UR - https://www.scopus.com/pages/publications/33747402940

U2 - 10.1016/b978-1-55860-335-6.50025-8

DO - 10.1016/b978-1-55860-335-6.50025-8

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:33747402940

T3 - Proceedings of the 11th International Conference on Machine Learning, ICML 1994

SP - 139

EP - 147

BT - Proceedings of the 11th International Conference on Machine Learning, ICML 1994

A2 - Cohen, William W.

A2 - Hirsh, Haym

PB - Morgan Kaufmann Publishers, Inc.

T2 - 11th International Conference on Machine Learning, ICML 1994

Y2 - 10 July 1994 through 13 July 1994

ER -

Getting the Most from Flawed Theories

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this