Domain-specific entity extraction from noisy, unstructured data using ontology-guided search

  • Sergey Bratus
  • Anna RumshiskyEmail author
  • Alexy Khrabrov
  • Rajenda Magar
  • Paul Thompson
Original Paper


Domain-specific knowledge is often recorded by experts in the form of unstructured text. For example, in the medical domain, clinical notes from electronic health records contain a wealth of information. Similar practices are found in other domains. The challenge we discuss in this paper is how to identify and extract part names from technicians repair notes, a noisy unstructured text data source from General Motors’ archives of solved vehicle repair problems, with the goal to develop a robust and dynamic reasoning system to be used as a repair adviser by service technicians. In the present work, we discuss two approaches to this problem. We present an algorithm for ontology-guided entity disambiguation that uses existing knowledge sources, such as domain-specific taxonomies and other structured data. We illustrate its use in the automotive domain, using GM parts ontology and the unit structure of repair manuals text to build context models, which are then used to disambiguate mentions of part-related entities in the text. We also describe extraction of part names with a small amount of annotated data using hidden Markov models (HMM) with shrinkage, achieving an f-score of approximately 80%. Next, we used linear-chain conditional random fields (CRF) in order to model observation dependencies present in the repair notes. Using CRF did not lead to improved performance, but a slight improvement over the HMM results was obtained by using a weighted combination of the HMM and CRF models.


Text analysis Language models Information extraction Ontology-guided search 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bruninghaus, S., Ashley, K.D.: Reasoning with Textual Cases Proceedings of the International Conference on Case-Based Reasoning (ICCBR), pp. 137–151 (2005)Google Scholar
  2. 2.
    Bundschus, M., Volker Tresp, V., Hans-Peter Kriegel, H.-P.: Topic models for semantically annotated document collections. In NIPS 2009 workshop: applications for topic models: text and beyond (2009)Google Scholar
  3. 3.
    Chapman, W., Dowling, J.N., Wagner, M.M.: Classification of emergency department chief complaints into 7 syndromes: a retrospective analysis of 527,228 patients. Annals of emergency medicine, vol. 46, no. 5, November (2005)Google Scholar
  4. 4.
    Cover T.M., Thomas J.A.: Elements of Information Theory. Wiley, New York (2006)zbMATHGoogle Scholar
  5. 5.
    Demner-Fushman, D.: UMLS content views appropriate for NLP processing of the biomedical literature vs. clinical text. J. Biomed. Inform. August (2010)Google Scholar
  6. 6.
    Fellbaum C. et al.: WordNet: An Electronic Lexical Database. MIT Press, New York (1998)zbMATHGoogle Scholar
  7. 7.
    Freitag, D., McCallum, A.: Information Extraction with HMM Structures Learned by Stochastic Optimization. In: Proceedings of the 17th National Conference on Artificial Intelligence, AAAI, pp. 584–589 (2000)Google Scholar
  8. 8.
    Freitag, D., McCallum, A.: Information Extraction with HMMs and Shrinkage. In: Papers from the AAAI-99 Workshop on Machine Learning for Information Extraction, 31–36, July. AAAI Technical Report WS-99-11 (1999)Google Scholar
  9. 9.
    Lafferty, J., McCallum, A., Pereira, F.: Conditional Random Fields: probabilistic Models for Segmenting and Labeling Sequence Data. In: Proceedings of 18th International Conference on Machine Learning (2001)Google Scholar
  10. 10.
    Lenz, M.: Textual CBR and information retreival: a comparison. In: Gierl L., Lenz M. (eds.) Proceedings of the 6th German Workshop on Case-Based Reasoning, IMIB Series vol. 7, Inst. fuer Medizinische Informatik und Biometrie, University of Rostock (1998)Google Scholar
  11. 11.
    Morgan A.P., Cafeo, J.A., Gibbons, D.I., Lesperance, R.M., Sengir, G.H., Simon, A.M.: The general motors variation-reduction adviser: evolution of a CBR system. ICCBR 2003, pp. 306–318 (2003)Google Scholar
  12. 12.
    Morgan A.P., Cafeo J.A., Godden K., Lesperance R.M., Simon A.M., McGuinness D.L., Benedict J.L.: The general motors variation-reduction adviser. AI Magazine 26(3), 18–28 (2005)Google Scholar
  13. 13.
    Rabiner, L.R.: A tutorial on Hidden Markov Models and selected applications in speech recognition. In: Proceedings of the IEEE, 77, 2 (1989)Google Scholar
  14. 14.
    Roberts, A., Gaizauskas, R., Hepple, M., Davis, N., Demetriou, G., Guo, Y., Kola, J., Roberts, I., Setzer, A., Tapuria, A. et al.: The CLEF corpus: semantic annotation of clinical text. In: AMIA Annual Symposium Proceedings, vol. 625 (2007)Google Scholar
  15. 15.
    Sha F., Pereira F.: Shallow Parsing with Conditional Random Fields Technical Report MS-CIS-02-35. University of Pennsylvania, Pennsylvania (2003)Google Scholar
  16. 16.
    Sutton C., McCallum A.: An introduction to conditional random fields for relational learning. In: Getoor, L., BenTaskar, B. (eds) Introduction to Statistical Relational Learning, MIT Press, New York (2006)Google Scholar
  17. 17.
    Uschold, M.: Creating, integrating and maintaining local and global ontologies. In: Proceedings of the 14th European Conference on Artificial Intelligence ECAI 2000, Berlin, Germany (2000)Google Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Sergey Bratus
    • 1
  • Anna Rumshisky
    • 3
    Email author
  • Alexy Khrabrov
    • 2
  • Rajenda Magar
    • 1
  • Paul Thompson
    • 1
  1. 1.Department of Computer ScienceDartmouth CollegeHanoverUSA
  2. 2.Thayer School of EngineeringDartmouth CollegeHanoverUSA
  3. 3.Department of Computer ScienceBrandeis UniversityWalthamUSA

Personalised recommendations