Skip to main content

Energy-Based Metric for Ensemble Selection

  • Conference paper
Web Technologies and Applications (APWeb 2012)

Part of the book series: Lecture Notes in Computer Science ((LNISA,volume 7235))

Included in the following conference series:

  • 2144 Accesses

Abstract

Ensemble selection copes with the reduction of an ensemble of the predictive models to reduce its response time and increase its accuracy. A number of selection methods via greedy search of the space of all possible ensemble subsets have been recently proposed. The major issue of these algorithms is to construct an effective metric to supervise the search process. In this paper, we view the issue of ensemble problem from a new viewpoint: energy-based learning, and then contribute a novel metric called EBM (Energy-based Metric) to guide the search. Also, this metric takes into account the strength of the decision of the current ensemble. Empirical results show that using the proposed metric to select subensemble leads to significantly better accuracy results compared to state-of-the-art metrics.

Supported by the National Science Foundation of China (No. 60901078).

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 84.99
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 109.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Kuncheva, L.I.: Combining Pattern Classifiers: Methods and Algorithms. John Wiley and Sons (2004)

    Google Scholar 

  2. Breiman, L.: Bagging predictors. Machine Learning, 123–140 (1996)

    Google Scholar 

  3. Freund, Y., Schapire, R.F.: A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences 55, 119–139 (1997)

    Article  MathSciNet  MATH  Google Scholar 

  4. Breiman, L.: Random forests. Machine learning 45, 5–32 (2001)

    Article  MATH  Google Scholar 

  5. Zhang, D., Chen, S., Zhou, Z., Yang, Q.: Constraint projections for ensemble learning. In: 23rd AAAI Conference on Artificial Intelligence (AAAI 2008), pp. 758–763 (2008)

    Google Scholar 

  6. Zhang, Y., Burer, S., Street, W.N.: Ensemble pruning via semi-definite programming. The Journal of Machine Learning Research, 1315–1338 (2006)

    Google Scholar 

  7. Chen, H., Tino, P., Yao, X.: Predictive ensemble pruning by expectation propagation. IEEE Transactions on Knowledge and Data Engineering, 999–1013 (2009)

    Google Scholar 

  8. Guo, H., Fan, M.: Ensemble Pruning via Base-Classifier Replacement. In: Wang, H., Li, S., Oyama, S., Hu, X., Qian, T. (eds.) WAIM 2011. LNCS, vol. 6897, pp. 505–516. Springer, Heidelberg (2011)

    Chapter  Google Scholar 

  9. Tamon, C., Xiang, J.: On the Boosting Pruning Problem. In: Lopez de Mantaras, R., Plaza, E. (eds.) ECML 2000. LNCS (LNAI), vol. 1810, pp. 404–412. Springer, Heidelberg (2000)

    Chapter  Google Scholar 

  10. Martinez-Muverbnoz, G., Suarez, A.: Aggregation ordering in bagging. In: International Conference on Artificial Intelligence and Applications (IASTED), pp. 258–263. Acta Press, Calgary (2004)

    Google Scholar 

  11. Banfield, R.E., Hall, L.O., Bowyer, K.W., Kegelmeyer, W.P.: Ensemble diversity measures and their application to thinning. Information Fusion, 49–62 (2005)

    Google Scholar 

  12. Partalas, I., Tsoumakas, G., Vlahavas, I.P.: Focused Ensemble Selection: A Diversity-Based Method for Greedy Ensemble Selection. In: 18th European Conference on Artificial Intelligence, pp. 117–121 (2008)

    Google Scholar 

  13. Martínez-Muñoz, G., Hernández-Lobato, D., Suárez, A.: An analysis of ensemble pruning techniques based on ordered aggregation. IEEE Transactions on Pattern Analysis and Machine Intelligence 31, 245–259 (2009)

    Article  Google Scholar 

  14. LeCun, Y., Chopra, S., Hadsell, R., Ranzato, M.A., Huang, F.J.: A Tutorial on Energy-Based Learning. In: Predicting Structured Data. MIT Press (2006)

    Google Scholar 

  15. Fan, W., Chu, F., Wang, H., Yu, P.S.: Pruning and dynamic scheduling of cost-sensitive ensembles. In: Eighteenth National Conference on Artificial Intelligence. American Association for Artificial Intelligence, pp. 146–151 (2002)

    Google Scholar 

  16. Partalas, I., Tsoumakas, G., Vlahavas, I.P.: An ensemble uncertainty aware measure for directed hill climbing ensemble pruning. Machine Learning, 257–282 (2010)

    Google Scholar 

  17. Asuncion, D.N.A.: UCI machine learning repository (2007)

    Google Scholar 

  18. Quinlan, J.R.: C4.5: programs for machine learning. Morgan Kaufmann (1993)

    Google Scholar 

  19. Witten, I.H., Frank, E.: Data Mining: Practical Machine Learning Tools and Techniques, 2nd edn. Morgan Kaufmann (2005)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2012 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Zhi, W., Guo, H., Fan, M. (2012). Energy-Based Metric for Ensemble Selection. In: Sheng, Q.Z., Wang, G., Jensen, C.S., Xu, G. (eds) Web Technologies and Applications. APWeb 2012. Lecture Notes in Computer Science, vol 7235. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-29253-8_26

Download citation

  • DOI: https://doi.org/10.1007/978-3-642-29253-8_26

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-642-29252-1

  • Online ISBN: 978-3-642-29253-8

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics