Advertisement

Preference Learning and Ranking by Pairwise Comparison

Chapter

Abstract

This chapter provides an overview of recent work on preference learning and ranking via pairwise classification. The learning by pairwise comparison (LPC) paradigm is the natural machine learning counterpart to the relational approach to preference modeling and decision making. From a machine learning point of view, LPC is especially appealing as it decomposes a possibly complex prediction problem into a certain number of learning problems of the simplest type, namely binary classification. We explain how to approach different preference learning problems, such as label and instance ranking, within the framework of LPC. We primarily focus on methodological aspects, but also address theoretical questions as well as algorithmic and complexity issues.

Keywords

Decision Boundary Binary Classifier Preference Learning Weighted Vote Relevant Label 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgements

This research has been supported by the German Science Foundation (DFG). We would like to thank our collaborators Klaus Brinker, Weiwei Cheng, Jens Hühn, Eneldo Loza Mencía, Sang-Hyeun Park, and Stijn Vanderlooy.

References

  1. 1.
    E.L. Allwein, R.E. Schapire, Y. Singer, Reducing multiclass to binary: A unifying approach for margin classifiers. J. Mach. Learn. Res. 1, 113–141 (2000)MathSciNetGoogle Scholar
  2. 2.
    C. Angulo, F.J. Ruiz, L. González, J.A. Ortega, Multi-classification by using tri-class SVM. Neural Process. Lett. 23(1), 89–101 (2006)CrossRefGoogle Scholar
  3. 3.
    R.A. Bradley, M.E. Terry, The rank analysis of incomplete block designs – I. The method of paired comparisons. Biometrika 39, 324–345 (1952)MathSciNetzbMATHGoogle Scholar
  4. 4.
    K. Brinker, J. Fürnkranz, E. Hüllermeier, A unified model for multilabel classification and ranking, in Proceedings of the 17th European Conference on Artificial Intelligence (ECAI-06), ed. by G.Brewka, S.Coradeschi, A.Perini, P.Traverso (2006), pp. 489–493Google Scholar
  5. 5.
    O. Dekel, C.D. Manning, Y. Singer, Log-linear models for label ranking, in Advances in Neural Information Processing Systems (NIPS-03), ed. by S. Thrun, L.K. Saul, B. Schölkopf (MIT, Cambridge, MA, 2004), pp. 497–504Google Scholar
  6. 6.
    J. Fodor, M. Roubens, Fuzzy Preference Modelling and Multicriteria Decision Support (Kluwer, 1994)Google Scholar
  7. 7.
    E. Frank, M. Hall, A simple approach to ordinal classification, in Proceedings of the 12th European Conference on Machine Learning (ECML-01), ed. by L.De Raedt, P.Flach (Springer, Freiburg, Germany, 2001), pp. 145–156Google Scholar
  8. 8.
    J.H. Friedman, Another approach to polychotomous classification. Technical report, Department of Statistics, Stanford University, Stanford, CA, 1996Google Scholar
  9. 9.
    J. Fürnkranz, Round robin classification. J. Mach. Learn. Res. 2, 721–747 (2002)MathSciNetzbMATHGoogle Scholar
  10. 10.
    J. Fürnkranz, Round robin ensembles. Intell. Data Anal. 7(5), 385–404 (2003)Google Scholar
  11. 11.
    J. Fürnkranz, E. Hüllermeier, Pairwise preference learning and ranking, in Proceedings of the 14th European Conference on Machine Learning (ECML-03), vol. 2837, Lecture Notes in Artificial Intelligence, ed. by N.Lavrač, D.Gamberger, H.Blockeel, L.Todorovski (Springer, Cavtat, Croatia, 2003), pp. 145–156Google Scholar
  12. 12.
    J. Fürnkranz, E. Hüllermeier, E. Loza Mencía, K. Brinker, Multilabel classification via calibrated label ranking. Mach. Learn. 73(2), 133–153 (2008)CrossRefGoogle Scholar
  13. 13.
    J. Fürnkranz, E. Hüllermeier, S. Vanderlooy, Binary decomposition methods for multipartite ranking, in Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD-09), vol. Part I, ed. by W.L. Buntine, M. Grobelnik, D. Mladenic, J. Shawe-Taylor (Springer, Bled, Slovenia, 2009), pp.359–374Google Scholar
  14. 14.
    S. Har-Peled, D. Roth, D. Zimak, Constraint classification: A new approach to multiclass classification, in Proceedings of the 13th International Conference on Algorithmic Learning Theory (ALT-02), ed. by N.Cesa-Bianchi, M.Numao, R.Reischuk (Springer, Lübeck, Germany, 2002), pp. 365–379CrossRefGoogle Scholar
  15. 15.
    T. Hastie, R. Tibshirani, Classification by pairwise coupling, in Advances in Neural Information Processing Systems 10 (NIPS-97), ed. by M.I. Jordan, M.J. Kearns, S.A. Solla (MIT, 1998), pp. 507–513Google Scholar
  16. 16.
    C.-W. Hsu, C.-J. Lin, A comparison of methods for multi-class support vector machines. IEEE Trans. Neural Netw. 13(2), 415–425 (2002)CrossRefGoogle Scholar
  17. 17.
    J. Hühn, E. Hüllermeier, Is an ordinal class structure useful in classifier learning? Intl. J. Data Min. Model. Manage. 1(1), 45–67 (2008)zbMATHCrossRefGoogle Scholar
  18. 18.
    J. Hühn, E. Hüllermeier, FR3: A fuzzy rule learner for inducing reliable classifiers. IEEE Trans. Fuzzy Syst. 17(1), 138–149 (2009)CrossRefGoogle Scholar
  19. 19.
    E. Hüllermeier, K. Brinker, Learning valued preference structures for solving classification problems. Fuzzy Sets Syst. 159(18), 2337–2352 (2008)zbMATHCrossRefGoogle Scholar
  20. 20.
    E. Hüllermeier, J. Fürnkranz, Comparison of ranking procedures in pairwise preference learning, in Proceedings of the 10th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU-04), Perugia, Italy, 2004Google Scholar
  21. 21.
    E. Hüllermeier, J. Fürnkranz, On predictive accuracy and risk minimization in pairwise label ranking. J. Comput. Syst. Sci. 76(1), 49–62 (2010)zbMATHCrossRefGoogle Scholar
  22. 22.
    E. Hüllermeier, J. Fürnkranz, W. Cheng, K. Brinker, Label ranking by learning pairwise preferences. Artif. Intell. 172, 1897–1916 (2008)zbMATHCrossRefGoogle Scholar
  23. 23.
    E. Hüllermeier, S. Vanderlooy, Combining predictions in pairwise classification: An optimal adaptive voting strategy and its relation to weighted voting. Pattern Recognit. 43(1), 128–142 (2010)Google Scholar
  24. 24.
    T. Joachims, Optimizing search engines using clickthrough data, in Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD-02), (ACM, 2002), pp. 133–142Google Scholar
  25. 25.
    S. Knerr, L. Personnaz, G. Dreyfus, Single-layer learning revisited: A stepwise procedure for building and training a neural network, in Neurocomputing: Algorithms, Architectures and Applications, vol. F68, NATO ASI Series, ed. by F.Fogelman Soulié, J.Hérault (Springer, 1990), pp. 41–50Google Scholar
  26. 26.
    S. Knerr, L. Personnaz, G. Dreyfus, Handwritten digit recognition by neural networks with single-layer training. IEEE Trans. Neural Netw. 3(6), 962–968 (1992)CrossRefGoogle Scholar
  27. 27.
    D. Koller, M. Sahami, Hierarchically classifying documents using very few words, in Proceedings of the 14th International Conference on Machine Learning (ICML-97) (Nashville, 1997), pp. 170–178Google Scholar
  28. 28.
    U.H.-G. Kreßel, Pairwise classification and support vector machines, in Advances in Kernel Methods: Support Vector Learning, Chap.15, ed. by B.Schölkopf, C.J.C. Burges, A.J. Smola (MIT, Cambridge, MA, 1999), pp. 255–268Google Scholar
  29. 29.
    E. Loza Mencía, J. Fürnkranz, Efficient pairwise multilabel classification for large-scale problems in the legal domain, in Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Disocvery in Databases (ECML-PKDD-2008), Part II, ed. by W. Daelemans, B. Goethals, K. Morik (Springer, Antwerp, Belgium, 2008), pp. 50–65Google Scholar
  30. 30.
    E. Loza Mencía, S.-H. Park, J. Fürnkranz, Efficient voting prediction for pairwise multilabel classification, in Proceedings of the 11th European Symposium on Artificial Neural Networks (ESANN-09) (d-side publicationsBruges, Belgium, 2009), pp. 117–122Google Scholar
  31. 31.
    B.-L. Lu, M. Ito, Task decomposition and module combination based on class relations: A modular neural network for pattern classification. IEEE Trans. Neural Netw. 10(5), 1244–1256 (1999)CrossRefGoogle Scholar
  32. 32.
    M. Moreira, E. Mayoraz, Improved pairwise coupling classification with correcting classifiers, in Proceedings of the 10th European Conference on Machine Learning (ECML-98), ed. by C.Nédellec, C.Rouveirol (Springer, Chemnitz, Germany, 1998), pp. 160–171CrossRefGoogle Scholar
  33. 33.
    G.H. Nam, Ordered pairwise classification. Master’s thesis, TU Darmstadt, Knowledge Engineering Group, 2007Google Scholar
  34. 34.
    S.-H. Park, J. Fürnkranz, Efficient pairwise classifciation, in Proceedings of 18th European Conference on Machine Learning (ECML-07), ed. by J.N. Kok, J.Koronacki, R.Lopez de Mantaras, S.Matwin, D.Mladenič, A.Skowron (Springer, Warsaw, Poland, 2007), pp.658–665Google Scholar
  35. 35.
    S.-H. Park, J. Fürnkranz, Efficient decoding of ternary error-correcting output codes for multiclass classification, in Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD-09), ed. by W.L.Buntine, M. Grobelnik, D. Mladenic, J. Shawe-Taylor, vol. Part I (Springer, Bled, Slovenia, 2009), pp. 189–204Google Scholar
  36. 36.
    J.C. Platt, N. Cristianini, J. Shawe-Taylor, Large margin DAGs for multiclass classification, in Advances in Neural Information Processing Systems 12 (NIPS-99), ed. by S.A. Solla, T.K.Leen, K.-R. Müller (MIT, 2000), pp. 547–553Google Scholar
  37. 37.
    D. Price, S. Knerr, L. Personnaz, G. Dreyfus, Pairwise neural network classifiers with probabilistic outputs, in Advances in Neural Information Processing Systems 7 (NIPS-94), ed. by G.Tesauro, D.Touretzky, T.Leen (MIT, 1995), pp. 1109–1116Google Scholar
  38. 38.
    T.L. Saaty, Decision Making for Leaders: The Analytic Hierarchy Process for Decisions in a Complex World (RWS Publications, Pittsburgh, Pennsylvania, 1999)Google Scholar
  39. 39.
    T.L. Saaty, Relative measurement and its generalization in decision making: Why pairwise comparisons are central in mathematics for the measurement of intangible factors – the analytic hierarchy/network process. Revista de la Real Academia de Ciencias Exactas Físicas y Naturales A Matemáticas (RACSAM) 102(2), 251–318 (2008)Google Scholar
  40. 40.
    M.S. Schmidt, H. Gish, Speaker identification via support vector classifiers, in Proceedings of the 21st IEEE International Conference Conference on Acoustics, Speech, and Signal Processing (ICASSP-96) (Atlanta, GA, 1996), pp. 105–108Google Scholar
  41. 41.
    J.F. Sima, Präferenz-Lernen für Hierarchische Multilabel Klassifikation. Master’s thesis, TU Darmstadt, Knowledge Engineering Group, 2007Google Scholar
  42. 42.
    J.-N. Sulzmann, J. Fürnkranz, E. Hüllermeier, On pairwise naive bayes classifiers, in Proceedings of 18th European Conference on Machine Learning (ECML-07), ed. by J.N. Kok, J.Koronacki, R.Lopez de Mantaras, S.Matwin, D.Mladenič, A.Skowron (Springer, Warsaw, Poland, 2007), pp. 371–381Google Scholar
  43. 43.
    L.L. Thurstone, A law of comparative judgement. Psychol. Rev. 34, 278–286 (1927)Google Scholar
  44. 44.
    T.-F. Wu, C.-J. Lin, R.C. Weng, Probability estimates for multi-class classification by pairwise coupling. J. Mach. Learn. Res. 5, 975–1005 (2004)MathSciNetzbMATHGoogle Scholar
  45. 45.
    Y. Yang, An evaluation of statistical approaches to text categorization. Inf. Retr. 1, 69–90 (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  1. 1.Technische Universität DarmstadtDarmstadtGermany
  2. 2.Philipps-Universität MarburgMarburgGermany

Personalised recommendations