Combining Event-Level and Cross-Event Semantic Information for Event-Oriented Relation Classification by SCNN

详细信息    查看全文

• 关键词：Event relation classification ; Semantic information ; Frame embedding ; SCNN
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：10035
• 期：1
• 页码：216-224
• 全文大小：282 KB
• 参考文献：1.Aharon, R.B., Szpektor, I., Dagan, I.: Generating entailment rules from framenet. In: Proceedings of ACL 2010 Conference Short Papers, pp. 241–246. Association for Computational Linguistics (2010)
  2.Baroni, M., Dinu, G., Kruszewski, G.: Don’t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors. In: ACL, vol. 1, pp. 238–247 (2014)
  3.Burchardt, A., Frank, A.: Approaching textual entailment with LFG and framenet frames. In: Proceedings of 2nd PASCAL RTE Challenge Workshop. Citeseer (2006)
  4.Chklovski, T., Pantel, P.: Global path-based refinement of noisy graphs applied to verb semantics. In: Dale, R., Wong, K.-F., Su, J., Kwong, O.Y. (eds.) IJCNLP 2005. LNCS (LNAI), vol. 3651, pp. 792–803. Springer, Heidelberg (2005)CrossRef
  5.Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., Kuksa, P.: Natural language processing (almost) from scratch. J. Mach. Learn. Res. 12, 2493–2537 (2011)MATH
  6.Ding, S., Hong, Y., Zhu, S., Yao, J., Zhu, Q.: Research of event relation classification based on tri-training. J. Comput. Eng. Sci. 37(12), 2345–2351 (2015)
  7.Fillmore, C.: Frame semantics. In: Linguistics in the Morning Calm, pp. 111–137 (1982)
  8.Fillmore, C.J., Johnson, C.R., Petruck, M.R.: Background to framenet. Int. J. Lexicogr. 16(3), 235–250 (2003)CrossRef
  9.Harris, Z.: Mathematical Structures of Language. Wiley, New York (1968)MATH
  10.Lin, D., Pantel, P.: Dirt@ sbt@ discovery of inference rules from text. In: Proceedings of 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 323–328. ACM (2001)
  11.Pantel, P., Pennacchiotti, M.: Espresso: leveraging generic patterns for automatically harvesting semantic relations. In: Proceedings of 21st International Conference on Computational Linguistics and 44th Annual Meeting of Association for Computational Linguistics, pp. 113–120. Association for Computational Linguistics (2006)
  12.Shen, D., Lapata, M.: Using semantic roles to improve question answering. In: EMNLP-CoNLL, pp. 12–21 (2007)
  13.Socher, R., Huang, E.H., Pennin, J., Manning, C.D., Ng, A.Y.: Dynamic pooling and unfolding recursive autoencoders for paraphrase detection. In: Advances in Neural Information Processing Systems, pp. 801–809 (2011)
  14.Yang, X., Hong, Y., Chen, Y., Wang, X., Yao, J., Zhu, Q.: Detection event relation through cross-scenario inference. J. Chin. Inf. Process. 28(5), 206–214 (2014)
  15.Zhang, B., Su, J., Xiong, D., Lu, Y., Duan, H., Yao, J.: Shallow convolutional neural network for implicit discourse relation recognition. In: Proceedings of 2015 Conference on Empirical Methods in Natural Language Processing, pp. 2230–2235 (2015)
  
• 作者单位：Siyuan Ding (18)
  Yu Hong (18)
  Shanshan Zhu (18)
  Jianmin Yao (18)
  Qiaoming Zhu (18)
  
  18. Provincial Key Laboratory of Computer Information Processing Technology, Soochow University, Suzhou, China
  
• 丛书名：Chinese Computational Linguistics and Natural Language Processing Based on Naturally Annotated Big Data
• ISBN：978-3-319-47674-2
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：10035

文摘

Previous researches on event relation classification primarily rely on lexical and syntactic features. In this paper, we use a Shallow Convolutional Neural Network (SCNN) to extract event-level and cross-event semantic features for event relation classification. On the one hand, the shallow structure alleviates the over-fitting problem caused by the lack of diverse relation samples. On the other hand, the utilization and combination of event-level and cross-event semantic information help improve relation classification. The experimental results show that our approach outperforms the state of the art.