Time-Evolving Relational Classification and Ensemble Methods

  • Ryan Rossi
  • Jennifer Neville
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7301)


Relational networks often evolve over time by the addition, deletion, and changing of links, nodes, and attributes. However, accurately incorporating the full range of temporal dependencies into relational learning algorithms remains a challenge. We propose a novel framework for discovering temporal-relational representations for classification. The framework considers transformations over all the evolving relational components (attributes, edges, and nodes) in order to accurately incorporate temporal dependencies into relational models. Additionally, we propose temporal ensemble methods and demonstrate their effectiveness against traditional and relational ensembles on two real-world datasets. In all cases, the proposed temporal-relational models outperform competing models that ignore temporal information.


Temporal Information Ensemble Method Attribute Weight Prediction Task Window Model 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bifet, A., Holmes, G., Pfahringer, B., Kirkby, R., Gavaldà, R.: New ensemble methods for evolving data streams. In: SIGKDD, pp. 139–148 (2009)Google Scholar
  2. 2.
    Chakrabarti, S., Dom, B., Indyk, P.: Enhanced hypertext categorization using hyperlinks. In: SIGMOD, pp. 307–318 (1998)Google Scholar
  3. 3.
    Cortes, C., Pregibon, D., Volinsky, C.: Communities of Interest. In: Hoffmann, F., Adams, N., Fisher, D., Guimarães, G., Hand, D.J. (eds.) IDA 2001. LNCS, vol. 2189, pp. 105–114. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  4. 4.
    Dietterich, T.G.: Ensemble Methods in Machine Learning. In: Kittler, J., Roli, F. (eds.) MCS 2000. LNCS, vol. 1857, pp. 1–15. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  5. 5.
    Domingos, P., Pazzani, M.: On the optimality of the simple bayesian classifier under zero-one loss. Machine Learning 29, 103–130 (1997)zbMATHCrossRefGoogle Scholar
  6. 6.
    Domingos, P., Richardson, M.: Mining the network value of customers. In: SIGKDD, pp. 57–66 (2001)Google Scholar
  7. 7.
    Dunlavy, D., Kolda, T., Acar, E.: Temporal link prediction using matrix and tensor factorizations. TKDD 5(2), 10 (2011)CrossRefGoogle Scholar
  8. 8.
    Eldardiry, H., Neville, J.: Across-model collective ensemble classification. AAAI (2011)Google Scholar
  9. 9.
    Güneş, İ., Çataltepe, Z., Öğüdücü, Ş.G.: GA-TVRC: A Novel Relational Time Varying Classifier to Extract Temporal Information Using Genetic Algorithms. In: Perner, P. (ed.) MLDM 2011. LNCS, vol. 6871, pp. 568–583. Springer, Heidelberg (2011)Google Scholar
  10. 10.
    Lahiri, M., Berger-Wolf, T.: Structure prediction in temporal networks using frequent subgraphs. In: CIDM, pp. 35–42 (2007)Google Scholar
  11. 11.
    McGovern, A., Collier, N., Matthew Gagne, I., Brown, D., Rodger, A.: Spatiotemporal Relational Probability Trees: An Introduction. In: ICDM, pp. 935–940 (2008)Google Scholar
  12. 12.
    Neville, J., Jensen, D., Friedland, L., Hay, M.: Learning relational probability trees. In: SIGKDD, pp. 625–630 (2003)Google Scholar
  13. 13.
    Neville, J., Jensen, D., Gallagher, B.: Simple estimators for relational Bayesian classifers. In: ICML, pp. 609–612 (2003)Google Scholar
  14. 14.
    Preisach, C., Schmidt-Thieme, L.: Relational ensemble classification. In: ICDM, pp. 499–509. IEEE (2006)Google Scholar
  15. 15.
    Preisach, C., Schmidt-Thieme, L.: Ensembles of relational classifiers. KIS 14(3), 249–272 (2008)zbMATHGoogle Scholar
  16. 16.
    Rossi, R., Neville, J.: Modeling the evolution of discussion topics and communication to improve relational classification. In: SOMA-KDD, pp. 89–97 (2010)Google Scholar
  17. 17.
    Rossi, R.A., Neville, J.: Representations and ensemble methods for dynamic relational classification. CoRR abs/1111.5312 (2011)Google Scholar
  18. 18.
    Sharan, U., Neville, J.: Temporal-relational classifiers for prediction in evolving domains. In: ICML (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Ryan Rossi
    • 1
  • Jennifer Neville
    • 1
  1. 1.Purdue UniversityWest LafayetteUSA

Personalised recommendations