Elamur Research
☰
home / papers / 2014 / The Stanford CoreNLP Natural Language Processing Toolkit

The Stanford CoreNLP Natural Language Processing Toolkit

Written for the Research library (claude-opus-5-5) from the paper text, in our own words: the license "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" does not allow redistributing the text itself. Original — aclanthology.org/P14-5010.pdf

Summary

Combining NLP components that each expose their own API usually requires fragile custom glue code. Large frameworks such as UIMA or GATE avoid that, but users have to learn the framework before they can analyse any text. This system-demonstration paper presents Stanford CoreNLP, a JVM-based toolkit that chains linguistic analysers into one simple pipeline. Input text goes into an Annotation object, a typesafe heterogeneous map, and a sequence of Annotators each add a layer to it. The layers run from tokenization and sentence splitting through POS tagging, lemmatization, NER and parsing to coreference and sentiment. Java Properties select and configure the annotators. A pipeline can run from a few lines of code or a single command and write its results as XML or plain text. The paper reports no benchmark scores. Instead, the authors credit the toolkit's wide use in research, industry and government to deliberate minimalism: few concepts to learn, plain Java objects, robust default models, a clear open-source license, documentation and responsiveness to users. The paper is a useful reference for anyone designing NLP infrastructure that is easy to adopt.

Key points

  • History: The pipeline was built for internal use in 2006. A multi-site grant project added a command-line interface and XML output in 2009, and the toolkit became open source under GPL v2+ in 2010. It is distributed via GitHub (stanfordnlp/CoreNLP) and Maven.
  • Annotators: tokenize (PTB-style, keeps character offsets), cleanxml, ssplit, truecase (CRF tagger), pos (maximum-entropy tagger), lemma, gender, ner, parse, dcoref and sentiment. The ner annotator covers PERSON, LOCATION, ORGANIZATION and MISC plus six numeric types: MONEY, NUMBER, DATE, TIME, DURATION and SET.
  • Languages: English models come bundled. Chinese and case-insensitive English are available as separate model packages, and French, German and Arabic are only partly supported.
  • Deliberate limits: The toolkit offers only a Java API and runs multi-threaded on a single machine. Spreading work across machines is left to systems such as Hadoop or Spark, unlike the heavier UIUC Curator.
  • Ecosystem: Third-party wrappers give access from Python, Ruby, Perl, Scala, Clojure, node.js and .NET. The toolkit can also run inside UIMA through wrappers such as ClearTK.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Named entity recognition · Part-of-speech tagging

Cited by

1582 papers of our library cite it (of the 140673 whose bibliographies we read); 80 of them have a page here, 1502 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/p14-5010

@inproceedings{manning2014stanford,
  title = {{The Stanford CoreNLP Natural Language Processing Toolkit}},
  author = {Manning, Christopher and Surdeanu, Mihai and Bauer, John and Finkel, Jenny and Bethard, Steven and McClosky, David},
  booktitle = {Proceedings of 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations},
  year = {2014},
  doi = {10.3115/v1/p14-5010},
  url = {https://doi.org/10.3115/v1/p14-5010}
}

References

16 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Steven Bethard, Philip Ogren, and Lee Becker. 2014. ClearTK 2.0: Design patterns for machine learning in UIMA. In LREC 2014. find
  2. Steven Bird, Ewan Klein, and Edward Loper. 2009. Natural Language Processing with Python. O’Reilly Media. find
  3. Joshua Bloch. 2008. Effective Java. Addison Wesley, Upper Saddle River, NJ, 2nd edition. find
  4. Angel X. Chang and Christopher D. Manning. 2012. SUTIME: A library for recognizing and normalizing time expressions. In LREC 2012. find
  5. James Clarke, Vivek Srikumar, Mark Sammons, and Dan Roth. 2012. An NLP Curator (or: How I learned to stop worrying and love NLP pipelines). In LREC 2012. find
  6. Hamish Cunningham, Diana Maynard, Kalina Bontcheva, and Valentin Tablan. 2002. GATE: an architecture for development of robust HLT applications. In ACL 2002. find
  7. Marie-Catherine de Marneffe, Bill MacCartney, and Christopher D. Manning. 2006. Generating typed dependency parses from phrase structure parses. In LREC 2006, pages 449–454. find our card
  8. David Ferrucci and Adam Lally. 2004. UIMA: an architectural approach to unstructured information processing in the corporate research environment. Natural Language Engineering, 10:327–348. find
  9. Jenny Rose Finkel, Trond Grenager, and Christopher Manning. 2005. Incorporating non-local information into information extraction systems by Gibbs sampling. In ACL 43, pages 363–370. find
  10. I. Gurevych, M. Mühlhäuser, C. Müller, J. Steimle, M. Weimer, and T. Zesch. 2007. Darmstadt knowledge processing repository based on UIMA. In First Workshop on Unstructured Information Management Architecture at GLDV 2007, Tübingen. find
  11. U. Hahn, E. Buyko, R. Landefeld, M. Mühlhausen, Poprat M, K. Tomanek, and J. Wermter. 2008. An overview of JCoRe, the Julie lab UIMA component registry. In LREC 2008. find
  12. Dan Klein and Christopher D. Manning. 2003. Fast exact inference with a factored model for natural language parsing. In Suzanna Becker, Sebastian Thrun, and Klaus Obermayer, editors, Advances in Neural Information Processing Systems, volume 15, pages 3–10. MIT Press. find
  13. Heeyoung Lee, Angel Chang, Yves Peirsman, Nathanael Chambers, Mihai Surdeanu, and Dan Jurafsky. 2013. Deterministic coreference resolution based on entity-centric, precision-ranked rules. Computational Linguistics, 39(4). find our card
  14. Anthony Patricio. 2009. Why this project is successful? https://community.jboss.org/wiki/WhyThisProjectIsSuccessful. ↗
  15. Richard Socher, Alex Perelygin, Jean Wu, Jason Chuang, Christopher D. Manning, Andrew Ng, and Christopher Potts. 2013. Recursive deep models for semantic compositionality over a sentiment treebank. In EMNLP 2013, pages 1631–1642. find
  16. Kristina Toutanova, Dan Klein, Christopher D. Manning, and Yoram Singer. 2003. Feature-rich part-of-speech tagging with a cyclic dependency network. In NAACL 3, pages 252–259. find