Advertisement

Cost Evaluation of CRF-Based Bibliography Extraction from Reference Strings

  • Naomichi Kawakami
  • Manabu Ohta
  • Atsuhiro Takasu
  • Jun Adachi
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8839)

Abstract

The effective use of digital libraries demands maintenance of bibliographic databases. Especially, the reference fields of academic papers are full of useful bibliographic information such as authors’ names and paper titles. We, therefore, propose a method of automatically extracting bibliographic information from reference strings using a conditional random field (CRF). However, at least a few hundred reference strings are necessary for training the CRF to achieve high extraction accuracies. As described herein, we propose the use of active sampling and pseudo-training data to reduce the amount of training data. Then we evaluate the associated training costs by experimentation.

Keywords

Information extraction CRF Reference string Active sampling Pseudo-training data 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Ohta, M., Arauchi, D., Takasu, A., Adachi, J.: Error detection of CRF-Based bibliography extraction from reference strings. In: Chen, H.-H., Chowdhury, G. (eds.) ICADL 2012. LNCS, vol. 7634, pp. 229–238. Springer, Heidelberg (2012)Google Scholar
  2. 2.
    Peng, F., McCallum, A.: Accurate information extraction from research papers using conditional random fields. In: HLT-NAACL, pp. 329–336 (2004)Google Scholar
  3. 3.
    Councill, I.G., Giles, C.L., Kan, M.Y.: ParsCit: An open-source CRF reference string parsing package. In: Proc. of Language Resources and Evaluation Conference (LREC 20), pp. 661–667 (2008)Google Scholar
  4. 4.
    Takasu, A., Ohta, M.: Rule management for information extraction from title pages of academic papers. In: Proc. of ICPRAM 2014, pp. 438–444 (2014)Google Scholar
  5. 5.
    Ohta, M., Arauchi, D., Takasu, A., Adachi, J.: Empirical evaluation of CRF-based bibliography extraction from reference strings. In: Proc. of IAPR DAS 2014, pp. 287–292 (2014)Google Scholar
  6. 6.
    Kudo, T., Yamamoto, K., Matsumoto, Y.: Applying conditional random fields to Japanese morphological analysis. In: Proc. of EMNLP 2004, pp. 230–237 (2004)Google Scholar
  7. 7.
    Settles, B., Craven, M.: An analysis of active learning strategies for sequence labeling tasks. In: Proc. of EMNLP 2008, pp. 1070–1079 (2008)Google Scholar
  8. 8.
    Saar-Tsechansky, M., Provost, F.: Active sampling for class probability estimation and ranking. Machine Learning 54, 153–178 (2004)CrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Naomichi Kawakami
    • 1
  • Manabu Ohta
    • 1
  • Atsuhiro Takasu
    • 2
  • Jun Adachi
    • 2
  1. 1.Okayama UniversityOkayamaJapan
  2. 2.National Instituteof InformaticsTokyoJapan

Personalised recommendations