Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
skip to main content
article
Free access

Support vector machine active learning with applications to text classification

Published: 01 March 2002 Publication History

Abstract

Support vector machines have met with significant success in numerous real-world learning tasks. However, like most machine learning algorithms, they are generally applied using a randomly selected training set classified in advance. In many settings, we also have the option of using <em>pool-based active learning</em>. Instead of using a randomly selected training set, the learner has access to a pool of unlabeled instances and can request the labels for some number of them. We introduce a new algorithm for performing active learning with support vector machines, i.e., an algorithm for choosing which instances to request next. We provide a theoretical motivation for the algorithm using the notion of a <em>version space</em>. We present experimental results showing that employing our active learning method can significantly reduce the need for labeled training instances in both the standard inductive and transductive settings.

References

[1]
C. J.C. Burges. A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 2:121-167, 1998.
[2]
C. Campbell, N. Cristianini, and A. Smola. Query learning with large margin classifiers. In Proceedings of the Seventeenth International Conference on Machine Learning, 2000.
[3]
G. Cauwenberghs and T. Poggio. Incremental and decremental support vector machine learning. In Advances in Neural Information Processing Systems, volume 13, 2001.
[4]
C. Cortes and V. Vapnik. Support vector networks. Machine Learning, 20:1-25, 1995.
[5]
I. Dagan and S. Engelson. Committee-based sampling for training probabilistic classifiers. In Proceedings of the Twelfth International Conference on Machine Learning, pages 150-157. Morgan Kaufmann, 1995.
[6]
S.T. Dumais, J. Platt, D. Heckerman, and M. Sahami. Inductive learning algorithms and representations for text categorization. In Proceedings of the Seventh International Conference on Information and Knowledge Management. ACM Press, 1998.
[7]
Y. Freund, H. Seung, E. Shamir, and N. Tishby. Selective sampling using the query by committee algorithm. Machine Learning, 28:133-168, 1997.
[8]
D. Heckerman, J. Breese, and K. Rommelse. Troubleshooting Under Uncertainty. Technical Report MSR-TR-94-07, Microsoft Research, 1994.
[9]
R. Herbrich, T. Graepel, and C. Campbell. Bayes point machines. Journal of Machine Learning Research, pages 245-279, 2001.
[10]
E. Horvitz and G. Rutledge. Time dependent utility and action under uncertainty. In Proceedings of the Seventh Conference on Uncertainty in Artificial Intelligence. Morgan Kaufmann, 1991.
[11]
T. Joachims. Text categorization with support vector machines. In Proceedings of the European Conference on Machine Learning. Springer-Verlag, 1998.
[12]
T. Joachims. Making large-scale svm learning practical. In B. Schölkopf, C. Burges, and A. Smola, editors, Advances in Kernel Methods - Support Vector Learning. MIT Press, 1999a.
[13]
T. Joachims. Transductive inference for text classification using support vector machines. In Proceedings of the Sixteenth International Conference on Machine Learning, pages 200-209. Morgan Kaufmann, 1999b.
[14]
K. Lang. Newsweeder: Learning to filter netnews. In International Conference on Machine Learning, pages 331-339, 1995.
[15]
Jean-Claude Latombe. Robot Motion Planning. Kluwer Academic Publishers, 1991.
[16]
D. Lewis and J. Catlett. Heterogeneous uncertainty sampling for supervised learning. In Proceedings of the Eleventh International Conference on Machine Learning, pages 148-156. Morgan Kaufmann, 1994.
[17]
D. Lewis and W. Gale. A sequential algorithm for training text classifiers. In Proceedings of the Seventeenth Annual International ACM-SIGIR Conference on Research and Development in Information Retrieval, pages 3-12. Springer-Verlag, 1994.
[18]
D. McAllester. PAC-Bayesian model averaging. In Proceedings of the Twelfth Annual Conference on Computational Learning Theory, 1999.
[19]
A. McCallum. Bow: A toolkit for statistical language modeling, text retrieval, classification and clustering, www.cs.cmu.edu/~mccallum/bow, 1996.
[20]
A. McCallum and K. Nigam. Employing EM in pool-based active learning for text classification. In Proceedings of the Fifteenth International Conference on Machine Learning. Morgan Kaufmann, 1998.
[21]
T. Mitchell. Generalization as search. Artificial Intelligence, 28:203-226, 1982.
[22]
J. Rocchio. Relevance feedback in information retrieval. In G. Salton, editor, The SMART retrieval system: Experiments in automatic document processing. Prentice-Hall, 1971.
[23]
G. Schohn and D. Cohn. Less is more: Active learning with support vector machines. In Proceedings of the Seventeenth International Conference on Machine Learning, 2000.
[24]
Fabrizio Sebastiani. Machine learning in automated text categorisation. Technical Report IEI-B4-31-1999, Istituto di Elaborazione dell'Informazione, 2001.
[25]
H.S. Seung, M. Opper, and H. Sompolinsky. Query by committee. In Proceedings of Computational Learning Theory, pages 287-294, 1992.
[26]
J. Shawe-Taylor and N. Cristianini. Further results on the margin distribution. In Proceedings of the Twelfth Annual Conference on Computational Learning Theory, pages 278-285, 1999.
[27]
V. Vapnik. Estimation of Dependences Based on Empirical Data. Springer Verlag, 1982.
[28]
V. Vapnik. Statistical Learning Theory. Wiley, 1998.

Cited By

View all
  • (2025)An Extensive Review of Machine Learning and Deep Learning Techniques on Network Intrusion Detection for IoTTransactions on Emerging Telecommunications Technologies10.1002/ett.7006436:2Online publication date: 14-Feb-2025
  • (2024)On the Pros and Cons of Active Learning for Moral Preference ElicitationProceedings of the 2024 AAAI/ACM Conference on AI, Ethics, and Society10.5555/3716662.3716724(711-723)Online publication date: 21-Oct-2024
  • (2024)Active statistical inferenceProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694680(62993-63010)Online publication date: 21-Jul-2024
  • Show More Cited By

Recommendations

Reviews

Brian Mayoh

Usability aspects of different visualizations of electronic documents are explored in this paper. To this end, it presents the results of an experiment studying the relationship between different electronic document interfaces and user reading activities, user efficiency in completing specific tasks, and overall user satisfaction. Hornbæk and Frøkjær compared three commonly used interfaces. The first is the linear interface, where the document is presented as a linear sequence of text and pictures. Next is the fisheye interface, in which the parts of the document that are considered important are always readable, while the remaining parts are initially collapsed, but may be expanded if users click on them. Third is the overview-and-detail interface, which consists of two panes: the overview pane, which includes section and subsection headings, and the detail pane, which displays the part of the document that is associated with the selected heading in the overview pane. The subjects of the experiment were students in the Computing Department at the University of Copenhagen. The documents used were scientific papers on topics relevant to the students' background. In three consecutive sessions, subjects were given three different documents, each in one of the three electronic formats described above. They were asked to read these documents and to complete specific essay and question-answering tasks. During each session, the user's reading behavior was monitored. Logged data on the user interactions was used to construct progression maps, depicting how the reading progressed, and visibility maps, showing the average length of time that different parts of the document were visible. At the end of each session, user performance with respect to the given tasks was measured. Finally, after having used all three interfaces, the user's subjective preference and satisfaction were recorded. Analysis of the reading behaviors of users revealed the existence of explicit reading patterns associated with each of the different electronic document interfaces. These reading patterns show how different interfaces affect the way people read documents. These reading patterns, in combination with the performance and satisfaction measurements obtained, were used to rationalize the usability aspects of different interfaces. The work presented in this paper is an elaborate survey, showing how electronic document interfaces affect reader's behavior and performance. It complements previous works on how interface designs influence user performance by explicitly identifying reading patterns, based on visual maps of reading activity, and by using these reading patterns to explain usability aspects of different visualizations of electronic documents. The results of the survey will benefit developers seeking to design interfaces that better support reading. Furthermore, the proposed visual maps of user reading activities can be further used by researchers concerned with the study and improvement of the visualization of electronic documents. Online Computing Reviews Service

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.

Comments

Information & Contributors

Information

Published In

cover image The Journal of Machine Learning Research
The Journal of Machine Learning Research  Volume 2, Issue
3/1/2002
735 pages
ISSN:1532-4435
EISSN:1533-7928
Issue’s Table of Contents

Publisher

JMLR.org

Publication History

Published: 01 March 2002
Published in JMLR Volume 2

Author Tags

  1. active learning
  2. classification
  3. relevance feedback
  4. selective sampling
  5. support vector machines

Qualifiers

  • Article

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)228
  • Downloads (Last 6 weeks)32
Reflects downloads up to 15 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2025)An Extensive Review of Machine Learning and Deep Learning Techniques on Network Intrusion Detection for IoTTransactions on Emerging Telecommunications Technologies10.1002/ett.7006436:2Online publication date: 14-Feb-2025
  • (2024)On the Pros and Cons of Active Learning for Moral Preference ElicitationProceedings of the 2024 AAAI/ACM Conference on AI, Ethics, and Society10.5555/3716662.3716724(711-723)Online publication date: 21-Oct-2024
  • (2024)Active statistical inferenceProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3694680(62993-63010)Online publication date: 21-Jul-2024
  • (2024)Data-efficient learning via clustering-based sensitivity samplingProceedings of the 41st International Conference on Machine Learning10.5555/3692070.3692153(2086-2107)Online publication date: 21-Jul-2024
  • (2024)Deep active learning with noise stabilityProceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence and Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence and Fourteenth Symposium on Educational Advances in Artificial Intelligence10.1609/aaai.v38i12.29270(13655-13663)Online publication date: 20-Feb-2024
  • (2024)Active learning guided by efficient surrogate learnersProceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence and Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence and Fourteenth Symposium on Educational Advances in Artificial Intelligence10.1609/aaai.v38i10.28961(10874-10881)Online publication date: 20-Feb-2024
  • (2024)Libby-Novick Beta-Liouville Distribution for Enhanced Anomaly Detection in Proportional DataACM Transactions on Intelligent Systems and Technology10.1145/367540515:5(1-26)Online publication date: 29-Jun-2024
  • (2024)Path Exploration Strategy for Symbolic Execution based on Multi-strategy Active LearningProceedings of the 15th Asia-Pacific Symposium on Internetware10.1145/3671016.3671403(165-168)Online publication date: 24-Jul-2024
  • (2024)Efficient Version Space Algorithms for Human-in-the-loop Model DevelopmentACM Transactions on Knowledge Discovery from Data10.1145/363744318:3(1-49)Online publication date: 12-Jan-2024
  • (2024)Say No to Redundant Information: Unsupervised Redundant Feature Elimination for Active LearningIEEE Transactions on Multimedia10.1109/TMM.2024.337119226(7721-7733)Online publication date: 18-Mar-2024
  • Show More Cited By

View Options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Login options

Full Access

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media