Abstract
Concept extraction is crucial for a number of downstream applications. However, surprisingly enough, straightforward single token/nominal chunk–concept alignment or dictionary lookup techniques such as DBpedia Spotlight still prevail. We propose a generic open domain-oriented extractive model that is based on distant supervision of a pointer–generator network leveraging bidirectional LSTMs and a copy mechanism and that is able to cope with the out-of-vocabulary phenomenon. The model has been trained on a large annotated corpus compiled specifically for this task from 250K Wikipedia pages, and tested on regular pages, where the pointers to other pages are considered as ground truth concepts. The outcome of the experiments shows that our model significantly outperforms standard techniques and, when used on top of DBpedia Spotlight, further improves its performance. The experiments furthermore show that the model can be readily ported to other datasets on which it equally achieves a state-of-the-art performance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We adopt Halliday’s notion of classifying nominal group as definition of a concept.
- 2.
- 3.
- 4.
We use a similar layout as in [30] for easier comparison of our extension with the original model.
- 5.
- 6.
Wikipedia does not contain self-links, therefore the concept “Grundy County” in a text from the self-titled page is not a link.
- 7.
Henceforth, we refer to the link snippet-based annotation of the pages as a sparse gold standard annotation since it covers by far not all concepts encountered in a page. Our distant supervision-based annotation is referred to as dense annotation since it (supposedly) covers all concepts on a given page. As usual, distant supervision-based annotation is also referred to as weak since it is an automatic annotation.
- 8.
- 9.
FRED [11] was not used as baseline as it is not scalable enough for the task: its REST service has a strong limitation on a number of possible requests per day, and it fails on processing long sentences (approximately more than 40 tokens).
- 10.
- 11.
- 12.
References
Akbik, A., Bergmann, T., Blythe, D., Rasul, K., Schweter, S., Vollgraf, R.: FLAIR: an easy-to-use framework for state-of-the-art NLP. In: Proceeding of NAACL (2019)
Al-Zaidy, R., Caragea, C., Giles, C.L.: Bi-LSTM-CRF sequence labeling for keyphrase extraction from scholarly documents. In: Proceeding of WWW (2019)
Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. In: 3rd International Conference on Learning Representations (2015)
Cheng, X., Roth, D.: Relational inference for wikification. In: Proceeding of the EMNLP, pp. 1787–1796 (2013)
Cordeiro, S., Ramisch, C., Villavicencio, A.: UFRGS&LIF at semeval-2016 task 10: rule-based MWE identification and predominant-supersense tagging. In: Proceeding of SemEval-2016, pp. 910–917 (2016)
Daiber, J., Jakob, M., Hokamp, C., Mendes, P.: Improving efficiency and accuracy in multilingual entity extraction. In: Proceeding of the 9th International Conference on Semantic Systems (I-Semantics) (2013)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: Pre-training of deep bidirectional transformers for language understanding. In: Proceeding of the NAACL-HLT, pp. 4171–4186 (2019)
Fader, A., Soderland, S., Etzioni, O.: Identifying relations for open information extraction. In: Proceeding of the EMNLP, pp. 1535–1545 (2011)
Fillmore, C., Baker, C.: Frame semantics for text understanding. In: Proceeding of the JWordNet and Other Lexical Resources Workshop at NAACL (2001)
Frege, G.: Ueber Sinn und Bedeutung. Zeitschrift fuer Philosophie und philosophische Kritik 100, 25–50 (1892)
Gangemi, A., Presutti, V., Reforgiato Recupero, D., Nuzzolese, A., Draicchio, F., Mongiovì, M.: Semantic web machine reading with fred. Semant. Web 8(6), 873–893 (2017)
Graves, A., Schmidhuber, J.: Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Netw. 18(5–6), 602–610 (2005)
Gu, J., Lu, Z., Li, H., Li, V.O.: Incorporating copying mechanism in sequence-to-sequence learning. In: Proceeding of the ACL, pp. 1631–1640 (2016)
Habibi, M., Weber, L., Neves, M., Wiegandt, D.L., Leser, U.: Deep learning with word embeddings improves biomedical named entity recognition. Bioinformatics 33(14), i37–i48 (2017)
Hailu, N.G.: Investigation of Traditional and Deep Neural Sequence Models for Biomedical Concept Recognition. Ph.D. thesis, University of Colorado (2019)
Halliday, M.: Halliday’s Introduction to Functional Grammar. Routledge, London & New York (2013)
Hasibi, F., Balog, K., Bratsberg, S.: Entity linking in queries: tasks and evaluation. In: Proceeding International Conference on The Theory of Information Retrieval, pp. 171–180. ACM (2015)
Honnibal, M., Montani, I.: spaCy 2: natural language understanding with bloom embeddings, convolutional neural networks and incremental parsing (2017). https://spacy.io/
Klein, G., Kim, Y., Deng, Y., Nguyen, V., Senellart, J., Rush, A.: Opennmt: neural machine translation toolkit. In: Proceeding of the 13th Conference of the AMTA, vol. 1, pp. 177–184 (2018)
Lample, G., Ballesteros, M., Subramanian, S., Kawakami, K., Dyer, C.: Neural architectures for named entity recognition. Proc. NAACL-HLT (2016)
Lin, Y., Michel, J.B., Aiden Lieberman, E., Orwant, J., Brockman, W., Petrov, S.: Syntactic annotations for the Google books NGram corpus. In: Proceeding of the ACL 2012 System Demonstrations, pp. 169–174, July 2012
Logeswaran, L., Chang, M.W., Lee, K., Toutanova, K., Devlin, J., Lee, H.: Zero-shot entity linking by reading entity descriptions. Proc. ACL, 3449–3460, July 2019
Luong, T., Pham, H., Manning, C.: Effective approaches to attention-based neural machine translation. Proc. EMNLP, 1412–1421 (2015)
Mausam, Schmitz, M., Soderland, S., Bart, R., Etzioni, O.: Open language learning for information extraction. In: Proceeding of the 2012 Joint EMNLP and CoNLL Conferences, pp. 523–534 (2012)
Meng, R., Zhao, S., Han, S., He, D., Brusilovsky, P., Chi, Y.: Deep keyphrase generation. Proc. ACL, 582–592 (2017)
Mintz, M., Bills, S., Snow, R., Jurafsky, D.: Distant supervision for relation extraction without labeled data. In: Proceeding of the ACL, pp. 1003–1011 (2009)
Petrucci, G., Rospocher, M., Ghidini, C.: Expressive ontology learning as neural machine translation. J. Web Semant. 52, 66–82 (2018)
Piccinno, F., Ferragina, P.: From TagME to WAT: a new entity annotator. In: Proceeding of the First International Workshop on Entity Recognition and Disambiguation, pp. 55–62. ERD ’14, ACM, New York, NY, USA (2014)
Schenkel, R., Suchanek, F., Kasneci, G.: Yawn: a semantically annotated wikipedia xml corpus. Datenbanksysteme in Business, Technologie und Web, -12 (2007)
See, A., Liu, P.J., Manning, C.D.: Get to the point: summarization with pointer-generator networks. Proc. ACL, 1073–1083 (2017)
Shang, J., Liu, J., Jiang, M., Ren, X., Voss, C., Han, J.: Automated phrase mining from massive text corpora. IEEE Trans. Knowl. Data Eng. 30(10), 1825–1837 (2018)
Straková, J., Straka, M., Hajic, J.: Neural architectures for nested NER through linearization. Proc. ACL, 5326–5331 (2019)
Tulkens, S., Šuster, S., Daelemans, W.: Unsupervised concept extraction from clinical text through semantic composition. J. Biomed. Inform. 91, 103–120 (2019)
Woods, W.A.: Conceptual Indexing: A Better Way to Organize Knowledge. Technical Report SMLI, TR97-61, Sun Microsystems Laboratories (1997)
Yosef, M., Hoffart, J., Bordino, I., Spaniol, M., Weikum, G.: Aida: an online tool for accurate disambiguation of named entities in text and tables. Proc. VLDB Endowment 4(12), 1450–1453 (2011)
Zhang, Z., Han, X., Liu, Z., Jiang, X., Sun, M., Liu, Q.: ERNIE: enhanced language representation with informative entities. Proc. ACL, 1441–1451 (2019)
Acknowledgments
The work presented in this paper has been supported by the European Commission within its H2020 Research Programme under the grant numbers 700024, 700475, 779962, 786731, 825079, and 870930.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Shvets, A., Wanner, L. (2020). Concept Extraction Using Pointer–Generator Networks and Distant Supervision for Data Augmentation. In: Keet, C.M., Dumontier, M. (eds) Knowledge Engineering and Knowledge Management. EKAW 2020. Lecture Notes in Computer Science(), vol 12387. Springer, Cham. https://doi.org/10.1007/978-3-030-61244-3_8
Download citation
DOI: https://doi.org/10.1007/978-3-030-61244-3_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-61243-6
Online ISBN: 978-3-030-61244-3
eBook Packages: Computer ScienceComputer Science (R0)