Abstract
Multi-label classification can be applied to study empirically discrete choice problems, in which each individual chooses more than one alternative. We applied the Classifier Chain (CC) method to transform the Generalized Maximum Entropy (GME) choice model from a single-label model to a multi-label model. The contribution of our CC-GME model lies in the advantages of both the GME and CC models. Specifically, the GME model can not only predict each individual’s choice, but also robustly estimate model parameters that describe factors determining his or her choices. The CC model is a problem transformation method that allows the decision on each alternative to be correlated. We used Monte-Carlo simulations and occupational hazard data to compare the CC-GME model with other selected methodologies for multi-label problems using the Hamming Loss, Accuracy, Precision and Recall measures. The results confirm the robustness of GME estimates with respect to relevant parameters regardless of the true error distributions. Moreover, the CC method outperforms other methods, indicating that the incorporation of the information on dependence patterns among alternatives can improve prediction performance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Baltas, G.: A model for multiple brand choice. Eur. J. Oper. Res. 154(1), 144–149 (2004)
Bhat, C.R., Srinivasan, S., Sen, S.: A joint model for the perfect and imperfect substitute goods case: application to activity time-use decisions. Transp. Res. Part B: Methodol. 40(10), 827–850 (2006)
Bhat, C.R., Srinivasan, S.: A multidimensional mixed ordered-response model for analyzing weekend activity participation. Transp. Res. Part B: Methodol. 39(3), 255–278 (2005)
Denzau, A.T., Gibbons, P.C., Greenberg, E.: Bayesian estimation of proportions with a cross-entropy prior. Commun. Stat.-Theory Methods 18(5), 1843–1861 (1989)
Golan, A., Judge, G., Perloff, J.M.: A maximum entropy approach to recovering information from multinomial response data. J. Am. Stat. Assoc. 91(434), 841–853 (1996)
Golan, A.: Information and Entropy Econometrics: A Review and Synthesis. Now Publishers Inc. (2008)
Heath, D., Zitzelberger, A., Giraud-Carrier, C.G.: A multiple domain comparison of multi-label classification methods. In: Working Notes of the 2nd International Workshop on Learning from Multi-label Data at ICML/COLT, 21–28 (2010)
Jaynes, E. T.: Information theory and statistical mechanics. Phys. rev. 106(4), 620 (1957a)
Jaynes, E. T.: Information theory and statistical mechanics. II. Phys. rev. 108(2), 171 (1957b)
Mosteller, F., Tukey, J.W.: Data Analysis, Including Statistics. The Collected Works of John W. Tukey: Graphics pp. 1965–1985, vol. 5 (123) (1988)
Pukelsheim, F.: The three sigma rule. Am. Stat. 48(2), 88–91 (1994)
Read, J., Pfahringer, B., Holmes, G., Frank, E.: Classifier chains for multi-label classification. Mach. Learn. 85(3), 333–359 (2011)
Santos, A., Canuto, A., Neto, A.F.: A comparative analysis of classification methods to multi-label tasks in different application domains. Int. J. Comput. Inform. Syst. Indust. Manag. Appl 3, 218–227 (2011)
Shannon, C.E.: A mathematical theory of communication. ACM SIGMOBILE Mob. Comput. Commun. Rev. 5(1), 3–55 (2001)
Soofi, E.S.: A generalizable formulation of conditional logit with diagnostics. J. Am. Stat. Assoc. 87, 812–816 (1992)
Train, K.: Data analysis. Including Statistics Discrete Choice Methods with Simulation. Cambridge University Press, Cambridge (2009)
Tsoumakas, G., Katakis, I.: Multi-label classification: an overview. Int. J. Data Warehous. Min. 3(3), 1–13 (2007)
Zhang, M.L., Zhou, Z.H.: ML-KNN: a lazy learning approach to multi-label learning. Pattern Recognit. 40(7), 2038–2048 (2007)
Acknowledgments
We are highly appreciated and would like to acknowledge the financial support from the Thailand Research Fund through the Royal Golden Jubilee Ph.D. Program (Grant No. PHD/0211/2556).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Leurcharusmee, S., Sirisrisakulchai, J., Sriboonchitta, S., Denœux, T. (2015). The Classifier Chain Generalized Maximum Entropy Model for Multi-label Choice Problems. In: Huynh, VN., Kreinovich, V., Sriboonchitta, S., Suriya, K. (eds) Econometrics of Risk. Studies in Computational Intelligence, vol 583. Springer, Cham. https://doi.org/10.1007/978-3-319-13449-9_13
Download citation
DOI: https://doi.org/10.1007/978-3-319-13449-9_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-13448-2
Online ISBN: 978-3-319-13449-9
eBook Packages: EngineeringEngineering (R0)