Skip to main content
Springer Nature Link
Log in

Non-Contextual vs Contextual Word Embeddings in Multiword Expressions Detection

  • Conference paper
  • First Online:
Computational Collective Intelligence (ICCCI 2022)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 13501))

Included in the following conference series:

  • 1256 Accesses

Abstract

Multiword Expression (MWE) detection is a crucial problem for many NLP applications. Recent methods approach it as a sequence labeling task and require manually annotated corpus. Traditional methods are based on statistical association measures and express limited accuracy, especially on smaller corpora. In this paper, we propose a novel weakly supervised method for extracting MWEs which concentrates on differences between interactions with context between the whole MWE and its component words. The interactions are represented by contextual embeddings (neural language models) and the observations are collected from various occurrence contexts of both the whole MWEs and their single word components. Our method uses a MWE lexicon as the sole knowledge base, and extracts training samples by matching the lexicon against a corpus to build classifiers for MWE recognition by Machine Learning. Thus, our approach does not require a corpus annotated with MWE occurrences, and also works with a limited corpus and a MWE list (\(\approx \)1400 MWEs in this work). It uses a general contextual embeddings model, HerBERTa, a kind of BERT model for Polish. The proposed method was evaluated on the Polish part of the PARSEME corpus and expressed very significant gain in comparison to the top methods from the PARSEME competition. The proposed method can be quite easily applied to other languages.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    Such MWEs form the vast majority of cases, both in PARSEME (only 111 out of 1,481 total correct MWE are longer) and plWordNet.

References

  1. Abadi, M., Agarwal, A., Barham, P., et al.: TensorFlow: large-scale machine learning on heterogeneous systems (2015). software available from tensorflow.org

    Google Scholar 

  2. Agrawal, S., Sanyal, R., Sanyal, S.: Hybrid method for automatic extraction of multiword expressions. Int. J. Eng. Technol. 7 (2018)

    Google Scholar 

  3. Anke, L.E., Schockaert, S., Wanner, L.: Collocation classification with unsupervised relation vectors. In: Proceedings of the 57th Annual Meeting of ACL, pp. 5765–5772 (2019)

    Google Scholar 

  4. Berk, G., Erden, B., Güngör, T.: Deep-BGT at PARSEME shared task 2018: bidirectional LSTM-CRF model for verbal multiword expression identification. In: LAW-MWE-CxG-2018, pp. 248–253. ACL (2018)

    Google Scholar 

  5. Bojanowski, P., Grave, E., Joulin, A., Mikolov, T.: Enriching word vectors with subword information. Trans. ACL 5, 135–146 (2017)

    Google Scholar 

  6. Boros, T., Burtica, R.: GBD-NER at PARSEME shared task 2018: multi-word expression detection using bidirectional long-short-term memory networks and graph-based decoding. In: Proceedings of the Joint Workshop on Linguistic Annotation, Multiword Expressions and Constructions, pp. 254–260. ACL (2018)

    Google Scholar 

  7. Broda, B., Derwojedowa, M., Piasecki, M.: Recognition of structured collocations in an inflective language. Syst. Sci. 34(4), 27–36 (2008)

    MATH  Google Scholar 

  8. Buljan, M., et al.: Combining linguistic features for the detection of Croatian multiword expressions. In: Proceedings of the 13th Workshop on MWE, pp. 194–199 (2017)

    Google Scholar 

  9. Chakraborty, S., Cougias, D., Piliero, S.: Identification of multiword expressions using transformers (2020). https://doi.org/10.13140/RG.2.2.31047.32169

  10. Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P.: Smote: synthetic minority over-sampling technique. J. AI Res. 16, 321–357 (2002)

    MATH  Google Scholar 

  11. Dziob, A., et al.: plWordNet 4.1-a linguistically motivated, corpus-based bilingual resource. In: Proceedings of the Tenth Global Wordnet Conference, pp. 353–362 (2019)

    Google Scholar 

  12. Evert, S.: The statistics of word cooccurrences: word pairs and collocations. Ph.D. thesis, Institut für maschinelle Sprachverarbeitung, Univ. of Stuttgart (2004)

    Google Scholar 

  13. Green, S., de Marneffe, M.C., Manning, C.D.: Parsing models for identifying multiword expressions. Comput. Linguist. 39(1), 195–227 (2013)

    Article  Google Scholar 

  14. Han, H., et al.: Borderline-smote: a new over-sampling method in imbalanced data sets learning. In: International Conference on Intelligent Computing, pp. 878–887 (2005)

    Google Scholar 

  15. Handler, A., Denny, M., Wallach, H., O’Connor, B.T.: Bag of what? simple noun phrase extraction for text analysis. In: NLP+CSS@EMNLP (2016)

    Google Scholar 

  16. He, H., Bai, Y., Garcia, E.A., Li, S.: ADASYN: adaptive synthetic sampling approach for imbalanced learning. In: 2008 IEEE International Joint Conference on Neural Networks, pp. 1322–1328. IEEE (2008)

    Google Scholar 

  17. Hosseini, M.J., et al.: UW-CSE at SemEval-2016 task 10: detecting multiword expressions and supersenses using double-chained conditional random fields. In: Proceedings of the 10th International Workshop on Semantic Evaluation, pp. 931–936 (2016)

    Google Scholar 

  18. Klyueva, N., Doucet, A., Straka, M.: Neural networks for multi-word expression detection. In: Proceedings of the 13th Workshop on MWE, pp. 60–65. ACL (2017)

    Google Scholar 

  19. Kocoń, J., Gawor, M.: Evaluating kgr10 polish word embeddings in the recognition of temporal expressions using bilstm-crf. Schedae Informaticae 27 (2018)

    Google Scholar 

  20. Kurfalı, M.: TRAVIS at PARSEME shared task 2020: how good is (m)BERT at seeing the unseen? In: Proceedings of the Joint Workshop on Multiword Expressions and Electronic Lexicons, pp. 136–141. ACL (2020)

    Google Scholar 

  21. Mikolov, T., Chen, K., Corrado, G., Dean, J.: Efficient estimation of word representations in vector space. In: Bengio, Y., LeCun, Y. (eds.) 1st International Conference on Learning Representations (2013)

    Google Scholar 

  22. Mroczkowski, R., Rybak, P., Wróblewska, A., Gawlik, I.: HerBERT: efficiently pretrained transformer-based language model for Polish. In: Proceedings of the 8th Workshop on Balto-Slavic Natural Language Processing, pp. 1–10. ACL (2021)

    Google Scholar 

  23. Pedregosa, F., Varoquaux, G., Gramfort, A., et al.: Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011)

    MathSciNet  MATH  Google Scholar 

  24. Pečina, P.: Lexical association measures and collocation extraction. Lang. Resour. Eval. 44, 137–158 (2010)

    Article  Google Scholar 

  25. Piasecki, M., Wendelberger, M., Maziarz, M.: Extraction of the multi-word lexical units in the perspective of the wordnet expansion. In: Proceedings of the International Conference Recent Advances in Natural Language Processing, pp. 512–520 (2015)

    Google Scholar 

  26. Ramisch, C.: Conclusions. In: Multiword Expressions Acquisition. TANLP, pp. 201–205. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-09207-2_8

    Chapter  Google Scholar 

  27. Ramisch, C., et al.: Edition 1.1 of the PARSEME shared task on automatic identification of verbal multiword expressions. In: Proceedings of the LAW-MWE-CxG-2018, pp. 222–240. ACL (2018)

    Google Scholar 

  28. Ramisch, C., et al.: Edition 1.2 of the PARSEME shared task on semi-supervised identification of verbal multiword expressions. In: Proceedings of the Joint Workshop on Multiword Expressions and Electronic Lexicons, pp. 107–118. ACL (2020)

    Google Scholar 

  29. Ramshaw, L., Marcus, M.: Text chunking using transformation-based learning. In: Third Workshop on Very Large Corpora (1995)

    Google Scholar 

  30. Rohanian, O., Taslimipoor, S., Kouchaki, S., Ha, L.A., Mitkov, R.: Bridging the gap: attending to discontinuity in identification of multiword expressions. In: Proceedings of the 2019 Conference of the North American Chapter of the ACL: Human Language Technologies, vol. 1, pp. 2692–2698. ACL (2019)

    Google Scholar 

  31. Saied, H.A., Candito, M., Constant, M.: Comparing linear and neural models for competitive MWE identification. In: Proceedings of the 22nd Nordic Conference on Computational Linguistics, pp. 86–96. Linköping University Electronic Press (2019)

    Google Scholar 

  32. Scholivet, M., Ramisch, C.: Identification of ambiguous multiword expressions using sequence models and lexical resources. In: Proceedings of the 13th Workshop on MWE, pp. 167–175 (2017)

    Google Scholar 

  33. Seretan, V.: Syntax-Based Collocation Extraction, Text, Speech and Language Technology, vol. 44. Springer, Netherlands (2011). https://doi.org/10.1007/978-94-007-0134-2

    Book  MATH  Google Scholar 

  34. Spasić, I., et al.: Unsupervised multi-word term recognition in Welsh. In: Proceedings of the Celtic Language Technology Workshop, pp. 1–6. EAMT (2019)

    Google Scholar 

  35. Taslimipoor, S., et al.: MTLB-STRUCT @parseme 2020: capturing unseen multiword expressions using multi-task learning and pre-trained masked language models. In: Proceedings of the Joint Workshop on Multiword Expressions and Electronic Lexicons, pp. 142–148 (2020)

    Google Scholar 

  36. Yirmibeşoğlu, Z., Güngör, T.: ERMI at PARSEME shared task 2020: embedding-rich multiword expression identification. In: Proceedings of the Joint Workshop on Multiword Expressions and Electronic Lexicons, pp. 130–135. ACL (2020)

    Google Scholar 

Download references

Acknowledgements

This work was partially supported by the National Science Centre, Poland, project no. 2019/33/B/HS2/02814; the statutory funds of the Department of Computational Intelligence, Wroclaw University of Science and Technology; the Polish Ministry of Education and Science, CLARIN-PL Project.

Author information

Authors and Affiliations

  1. Department of Artificial Intelligence, Wrocław University of Science and Technology, Wrocław, Poland

    Maciej Piasecki & Kamil Kanclerz

Authors
  1. Maciej Piasecki
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kamil Kanclerz
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Maciej Piasecki .

Editor information

Editors and Affiliations

  1. Wrocław University of Science and Technology, Wrocław, Poland

    Ngoc Thanh Nguyen

  2. Open University of Cyprus, Nicosia, Cyprus

    Yannis Manolopoulos

  3. University of Pau and Pays de l'Adour, Anglet, France

    Richard Chbeir

  4. Wrocław University of Science and Technology, Wrocław, Poland

    Adrianna Kozierkiewicz

  5. Wrocław University of Science and Technology, Wrocław, Poland

    Bogdan Trawiński

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Piasecki, M., Kanclerz, K. (2022). Non-Contextual vs Contextual Word Embeddings in Multiword Expressions Detection. In: Nguyen, N.T., Manolopoulos, Y., Chbeir, R., Kozierkiewicz, A., Trawiński, B. (eds) Computational Collective Intelligence. ICCCI 2022. Lecture Notes in Computer Science(), vol 13501. Springer, Cham. https://doi.org/10.1007/978-3-031-16014-1_16

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-16014-1_16

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-16013-4

  • Online ISBN: 978-3-031-16014-1

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics