Elamur Research⌕
☰
home / papers / 2014 / Frame Semantic Tree Kernels for Social Network Extraction from Text

Frame Semantic Tree Kernels for Social Network Extraction from Text

Summary

The paper addresses automatic extraction of social networks from text. Nodes are person entities, and links are social events of two kinds: mutual interactions (INR) and one-sided observations (OBS), where one person thinks or talks about another. Building on their earlier ACE2005 social-event annotation and the syntactic tree-kernel system AR2010, the authors add semantic information from FrameNet frame parses produced by Semafor. They test two kinds of semantic input. The first is bag-of-frames vectors and hand-written frame-based rules. The second is three new tree representations of frame parses (FrameForest, FrameTree, FrameTreeProp), used inside SVM convolution kernels. They also evaluate the full network extraction task by merging detection and classification, comparing One-versus-All against a hierarchical pipeline. Tree kernels over semantic structures proved more useful than hand-crafted rules, because they tolerate partial or erroneous parses. The best system combined lexical, syntactic, and semantic representations, and the hierarchical design was significantly better. This matters for analysts who need interaction networks from large text collections in fields such as history, literature, and journalism.

Key points

  • Data: ACE2005 with social event annotation, containing 199 INR, 199 OBS, and 1,609 No-Event entity pairs. Gold named entity mentions were used.
  • Semantic features: Semafor found 1,174 distinct FrameNet frames in the corpus. The authors wrote 199 rules from the FrameNet manual, but these fired in only 17% of true events (versus 7% of non-events) and gave no overall gain.
  • Best combination: AR2010 plus FrameTreeProp raised social event detection F1 from 0.574 to 0.585. Classification accuracy rose from 81.1% to 82.0%, with FrameForest reaching 82.6%.
  • Full task: For the three-way social network extraction task, hierarchical F1 improved from 0.465 to 0.480, a statistically significant gain.
  • Pipeline design: The hierarchical design (detect first, then classify) beat One-versus-All for every model by a significant margin.

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: Semantic role labeling

Cited by

15 papers of our library cite it (of the 140673 whose bibliographies we read); 2 of them have a page here, 13 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/e14-1023

@inproceedings{agarwal2014frame,
  title = {{Frame Semantic Tree Kernels for Social Network Extraction from Text}},
  author = {Agarwal, Apoorv and Balasubramanian, Sriramkumar and Kotalwar, Anup and Zheng, Jiehan and Rambow, Owen},
  booktitle = {Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics},
  year = {2014},
  doi = {10.3115/v1/e14-1023},
  url = {https://doi.org/10.3115/v1/e14-1023}
}

References

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

  1. Apoorv Agarwal and Owen Rambow. 2010. Automatic detection and classification of social events. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 1024–1034, Cambridge, MA, October. Association for Computational Linguistics. find
  2. Apoorv Agarwal, Owen C. Rambow, and Rebecca J. Passonneau. 2010. Annotation scheme for social network extraction from text. In Proceedings of the Fourth Linguistic Annotation Workshop. find
  3. Apoorv Agarwal, Anup Kotalwar, and Owen Rambow. 2013a. Automatic extraction of social networks from literary text: A case study on alice in wonderland. In the Proceedings of the 6th International Joint Conference on Natural Language Processing (IJCNLP 2013). find
  4. Apoorv Agarwal, Anup Kotalwar, Jiehan Zheng, and Owen Rambow. 2013b. Sinnet: Social interaction network extractor from text. In Sixth International Joint Conference on Natural Language Processing, page 33. find
  5. Collin F. Baker, J. Fillmore, and John B. Lowe. 1998. The Berkeley FrameNet project. In 36th Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics (COLING-ACL’98), pages 86–90, Montréal. find our card
  6. Chris Burges. 1998. A tutorial on support vector machines for pattern recognition. Data mining and knowledge discovery. find
  7. Desai Chen, Nathan Schneider, Dipanjan Das, and Noah A. Smith. 2010. Semafor: Frame argument resolution with log-linear models. In Proceedings of the 5th International Workshop on Semantic Evaluation, pages 264–267, Uppsala, Sweden, July. Association for Computational Linguistics. find
  8. Michael Collins and Nigel Duffy. 2002. New ranking algorithms for parsing and tagging: Kernels over discrete structures, and the voted perceptron. In Proceedings of the 40th annual meeting on association for computational linguistics, pages 263–270. Association for Computational Linguistics. find
  9. Aron Culotta and Jeffrey Sorensen. 2004. Dependency tree kernels for relation extraction. In Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL’04), Main Volume, pages 423–429, Barcelona, Spain, July. find
  10. David K. Elson, Nicholas Dames, and Kathleen R. McKeown. 2010. Extracting social networks from literary fiction. Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, pages 138–147. find
  11. Zhou GuoDong, Su Jian, Zhang Jie, and Zhang Min. 2005. Exploring various knowledge in relation extraction. In Proceedings of 43th Annual Meeting of the Association for Computational Linguistics. find
  12. Sanda Harabagiu, Cosmin Adrian Bejan, and Paul Morarescu. 2005. Shallow semantics for relation extraction. In International Joint Conference On Artificial Intelligence. find
  13. David Haussler. 1999. Convolution kernels on discrete structures. Technical report, University of California at Santa Cruz. find
  14. Hua He, Denilson Barbosa, and Grzegorz Kondrak. 2013. Identification of speakers in novels. The 51st Annual Meeting of the Association for Computational Linguistics (ACL 2013). find
  15. Nanda Kambhatla. 2004. Combining lexical, syntactic, and semantic features with maximum entropy models for extracting relations. In Proceedings of the ACL 2004 on Interactive poster and demonstration sessions, page 22. Association for Computational Linguistics. find
  16. Dan Klein and Christopher D. Manning. 2003. Accurate unlexicalized parsing. Proceedings of the 41st Meeting of the Association for Computational Linguistics, pages 423–430. find our card
  17. Alessandro Moschitti. 2006. Efficient convolution kernels for dependency and constituent syntactic trees. In Proceedings of the 17th European Conference on Machine Learning. find
  18. Truc-Vien T. Nguyen, Alessandro Moschitti, and Giuseppe Riccardi. 2009. Convolution kernels on constituent, dependency and sequential structures for relation extraction. Conference on Empirical Methods in Natural Language Processing. find
  19. Barbara Plank and Alessandro Moschitti. 2013. Embedding semantic similarity in tree kernels for domain adaptation of relation extraction. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1498–1507, Sofia, Bulgaria, August. Asso… find
  20. SVN Vishwanathan, Nicol N Schraudolph, Risi Kondor, and Karsten M Borgwardt. 2010. Graph kernels. The Journal of Machine Learning Research, 11:1201–1242. find
  21. Dmitry Zelenko, Chinatsu Aone, and Anthony Richardella. 2003. Kernel methods for relation extraction. The Journal of Machine Learning Research, 3:1083–1106. find
  22. Shubin Zhao and Ralph Grishman. 2005. Extracting relations with integrated information using kernel methods. In Proceedings of the 43rd Meeting of the ACL. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/agarwal-2014-frame-semantic-tree-kernels-social/