Advertisement

Unsupervised Learning of Link Discovery Configuration

  • Andriy Nikolov
  • Mathieu d’Aquin
  • Enrico Motta
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7295)

Abstract

Discovering links between overlapping datasets on the Web is generally realised through the use of fuzzy similarity measures. Configuring such measures is often a non-trivial task that depends on the domain, ontological schemas, and formatting conventions in data. Existing solutions either rely on the user’s knowledge of the data and the domain or on the use of machine learning to discover these parameters based on training data. In this paper, we present a novel approach to tackle the issue of data linking which relies on the unsupervised discovery of the required similarity parameters. Instead of using labeled data, the method takes into account several desired properties which the distribution of output similarity values should satisfy. The method includes these features into a fitness criterion used in a genetic algorithm to establish similarity parameters that maximise the quality of the resulting linkset according to the considered properties. We show in experiments using benchmarks as well as real-world datasets that such an unsupervised method can reach the same levels of performance as manually engineered methods, and how the different parameters of the genetic algorithm and the fitness criterion affect the results for different datasets.

Keywords

Genetic Algorithm Decision Rule Candidate Solution Unsupervised Learn Ontology Match 
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.
    de Carvalho, M.G., Laender, A.H.F., Goncalves, M.A., da Silva, A.S.: A genetic programming approach to record deduplication. IEEE Transactions on Knowledge and Data Engineering 99(PrePrints) (2010)Google Scholar
  2. 2.
    Chaudhuri, S., Ganti, V., Motwani, R.: Robust identification of fuzzy duplicates. In: ICDE 2005, pp. 865–876 (2005)Google Scholar
  3. 3.
    Elmagarmid, A.K., Ipeirotis, P.G., Verykios, V.S.: Duplicate record detection: A survey. IEEE Transactions on Knowledge and Data Engineering 19(1), 1–16 (2007)CrossRefGoogle Scholar
  4. 4.
    Euzenat, J., et al.: Results of the ontology alignment evaluation initiative 2010. In: Workshop on Ontology Matching (OM 2010), ISWC 2010 (2010)Google Scholar
  5. 5.
    Euzenat, J., et al.: Results of the ontology alignment evaluation initiative 2011. In: Workshop on Ontology Matching (OM 2011), ISWC 2011 (2011)Google Scholar
  6. 6.
    Fellegi, I.P., Sunter, A.B.: A theory for record linkage. Journal of American Statistical Association 64(328), 1183–1210 (1969)Google Scholar
  7. 7.
    Hu, W., Chen, J., Qu, Y.: A self-training approach for resolving object coreference on the semantic web. In: WWW 2011, pp. 87–96 (2011)Google Scholar
  8. 8.
    Isele, R., Bizer, C.: Learning linkage rules using genetic programming. In: Workshop on Ontology Matching (OM 2011), ISWC 2011, Bonn, Germany (2011)Google Scholar
  9. 9.
    Li, J., Tang, J., Li, Y., Luo, Q.: RiMOM: A dynamic multistrategy ontology alignment framework. IEEE Transactions on Knowledge and Data Engineering 21(8), 1218–1232 (2009)CrossRefGoogle Scholar
  10. 10.
    Ngonga Ngomo, A.C., Lehmann, J., Auer, S., Höffner, K.: RAVEN - active learning of link specifications. In: Workshop on Ontology Matching (OM 2011), ISWC 2011 (2011)Google Scholar
  11. 11.
    Nikolov, A., Uren, V., Motta, E., de Roeck, A.: Integration of Semantically Annotated Data by the KnoFuss Architecture. In: Gangemi, A., Euzenat, J. (eds.) EKAW 2008. LNCS (LNAI), vol. 5268, pp. 265–274. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  12. 12.
    Noessner, J., Niepert, M., Meilicke, C., Stuckenschmidt, H.: Leveraging Terminological Structure for Object Reconciliation. In: Aroyo, L., Antoniou, G., Hyvönen, E., ten Teije, A., Stuckenschmidt, H., Cabral, L., Tudorache, T. (eds.) ESWC 2010. LNCS, vol. 6089, pp. 334–348. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  13. 13.
    Stoermer, H., Rassadko, N., Vaidya, N.: Feature-Based Entity Matching: The FBEM Model, Implementation, Evaluation. In: Pernici, B. (ed.) CAiSE 2010. LNCS, vol. 6051, pp. 180–193. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  14. 14.
    Volz, J., Bizer, C., Gaedke, M., Kobilarov, G.: Discovering and Maintaining Links on the Web of Data. In: Bernstein, A., Karger, D.R., Heath, T., Feigenbaum, L., Maynard, D., Motta, E., Thirunarayan, K. (eds.) ISWC 2009. LNCS, vol. 5823, pp. 650–665. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  15. 15.
    Zardetto, D., Scannapietro, M., Catarci, T.: Effective automated object matching. In: ICDE 2010, pp. 757–768 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Andriy Nikolov
    • 1
  • Mathieu d’Aquin
    • 1
  • Enrico Motta
    • 1
  1. 1.Knowledge Media InstituteThe Open UniversityMilton KeynesUK

Personalised recommendations