Advertisement

Network Regression with Predictive Clustering Trees

  • Daniela Stojanova
  • Michelangelo Ceci
  • Annalisa Appice
  • Sašo Džeroski
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6913)

Abstract

Regression inference in network data is a challenging task in machine learning and data mining. Network data describe entities represented by nodes, which may be connected with (related to) each other by edges. Many network datasets are characterized by a form of autocorrelation where the values of the response variable at a given node depend on the values of the variables (predictor and response) at the nodes connected to the given node. This phenomenon is a direct violation of the assumption of independent (i.i.d.) observations: At the same time, it offers a unique opportunity to improve the performance of predictive models on network data, as inferences about one entity can be used to improve inferences about related entities. In this paper, we propose a data mining method that explicitly considers autocorrelation when building regression models from network data. The method is based on the concept of predictive clustering trees (PCTs), which can be used both for clustering and predictive tasks: PCTs are decision trees viewed as hierarchies of clusters and provide symbolic descriptions of the clusters. In addition, PCTs can be used for multi-objective prediction problems, including multi-target regression and multi-target classification. Empirical results on real world problems of network regression show that the proposed extension of PCTs performs better than traditional decision tree induction when autocorrelation is present in the data.

Keywords

Root Mean Square Error Network Data Variance Reduction Average Mean Square Error Descriptive Space 
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.

References

  1. 1.
    Angin, P., Neville, J.: A shrinkage approach for modeling non-stationary relational autocorrelation. In: Proc. 8th IEEE Intl. Conf. on Data Mining, pp. 707–712 (2008)Google Scholar
  2. 2.
    Appice, A., Ceci, M., Malerba, D.: An iterative learning algorithm for within-network regression in the transductive setting. In: Discovery Science, pp. 36–50 (2009)Google Scholar
  3. 3.
    Blockeel, H., De Raedt, L., Ramon, J.: Top-down induction of clustering trees. In: Proc. 15th Intl. Conf. on Machine Learning, pp. 55–63 (1998)Google Scholar
  4. 4.
    Breiman, L., Friedman, J., Olshen, R., Stone, J.: Classification and Regression trees. Wadsworth & Brooks, Belmont (1984)zbMATHGoogle Scholar
  5. 5.
    Brent, R.: Algorithms for Minimization without Derivatives. Prentice-Hall, Englewood Cliffs (1973)zbMATHGoogle Scholar
  6. 6.
    Cortez, P., Morais, A.: A Data Mining Approach to Predict Forest Fires using Meteorological Data. In: Proc. 13th Portuguese Conf. Artificial Intelligence, New Trends in Artificial Intelligence, pp. 512–523 (2007)Google Scholar
  7. 7.
    Demšar, D., Debeljak, M., Lavigne, C., Džeroski S.: Modelling pollen dispersal of genetically modified oilseed rape within the field. In: Abstracts of the 90th ESA Annual Meeting, p. 152. The Ecological Society of America (2005)Google Scholar
  8. 8.
    Džeroski, S., Gjorgjioski, V., Slavkov, I., Struyf, J.: Analysis of time series data with predictive clustering trees. In: Proc. 5th Intl. Wshp. on Knowledge Discovery in Inductive Databases, pp. 63–80. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  9. 9.
    Fotheringham, A.S., Brunsdon, C., Charlton, M.: Geographically Weighted Regression: The Analysis of Spatially Varying Relationships. Wiley, Chichester (2002)zbMATHGoogle Scholar
  10. 10.
    Gallagher, B., Tong, H., Eliassi-Rad, T., Faloutsos, C.: Using ghost edges for classification in sparsely labeled networks. In: Proc. 14th ACM SIGKDD Intl. Conf. Knowledge Discovery and Data Mining, pp. 256–264 (2008)Google Scholar
  11. 11.
    Góra, G., Wojna, A.: RIONA: A classifier combining rule induction and k-NN method with automated selection of optimal neighbourhood. In: Elomaa, T., Mannila, H., Toivonen, H. (eds.) ECML 2002. LNCS (LNAI), vol. 2430, pp. 111–123. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  12. 12.
    Jensen, D., Neville, J., Gallagher, B.: Why collective inference improves relational classification. In: Proc. 10th Intl. Conf. on Knowledge Discovery and Data Mining, pp. 593–598 (2004)Google Scholar
  13. 13.
    Legendre, P.: Spatial autocorrelation: Trouble or new paradigm? Ecology 74(6), 1659–1673 (1993)CrossRefGoogle Scholar
  14. 14.
    Macskassy, S., Provost, F.: Classification in networked data: a toolkit and a univariate case study. Machine Learning 8, 935–983 (2007)Google Scholar
  15. 15.
    Macskassy, S.A.: Improving learning in networked data by combining explicit and mined links. In: Proc. 22th Intl. Conf. on Artificial Intelligence, pp. 590–595 (2007)Google Scholar
  16. 16.
    McPherson, M., Smith-Lovin, L., Cook, J.: Birds of a feather: Homophily in social networks. Annual Review of Sociology 27, 415–444 (2001)CrossRefGoogle Scholar
  17. 17.
    Mehta, M., Agrawal, R., Rissanen, J.: Sliq: A fast scalable classifier for data mining. In: Apers, P.M.G., Bouzeghoub, M., Gardarin, G. (eds.) EDBT 1996. LNCS, vol. 1057, pp. 18–32. Springer, Heidelberg (1996)CrossRefGoogle Scholar
  18. 18.
    Michalski, R.S., Stepp, R.: Machine Learning: An Artificial Intelligence Approach. In: Learning from Observation: Conceptual Clustering, Tioga, pp. 331–363 (2003)Google Scholar
  19. 19.
    Neville, J., Jensen, D.: Relational dependency networks. Journal of Machine Learning Research 8, 653–692 (2007)zbMATHGoogle Scholar
  20. 20.
    Neville, J., Simsek, O., Jensen, D.: Autocorrelation and relational learning: Challenges and opportunities. In: Wshp. Statistical Relational Learning (2004)Google Scholar
  21. 21.
    Orkin, M., Drogin, R.: Vital Statistics. McGraw Hill, New York (1990)zbMATHGoogle Scholar
  22. 22.
    Pace, P., Barry, R.: Quick computation of regression with a spatially autoregressive dependent variable. Geographical Analysis 29(3), 232–247 (1997)CrossRefGoogle Scholar
  23. 23.
    Randic, M.: On characterization of molecular attributes. Journal of American Chemical Society (1975)Google Scholar
  24. 24.
    Sen, P., Namata, G., Bilgic, M., Getoor, L., Gallagher, B., Eliassi-Rad, T.: Collective classification in network data. AI Magazine 29(3), 93–106 (2008)Google Scholar
  25. 25.
    Vapnik, V.: Statistical Learning Theory. Wiley, New York (1998)zbMATHGoogle Scholar
  26. 26.
    Zhu, X., Ghahramani, Z., Lafferty, J.D.: Semi-supervised learning using gaussian fields and harmonic functions. In: Proc. 20th Intl. Conf. on Machine Learning, pp. 912–919 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Daniela Stojanova
    • 1
  • Michelangelo Ceci
    • 2
  • Annalisa Appice
    • 2
  • Sašo Džeroski
    • 1
  1. 1.Department of Knowledge TechnologiesJožef Stefan InstituteLjubljanaSlovenia
  2. 2.Dipartimento di InformaticaUniversità degli Studi di Bari “Aldo Modo”BariItaly

Personalised recommendations