Research on Pattern Representation and Reliability in Semi-Supervised Entity Relation Extraction

详细信息    查看全文

• 关键词：Entity relation ; Semi ; supervised ; Semantic pattern ; Reliability
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9713
• 期：1
• 页码：289-297
• 全文大小：316 KB
• 参考文献：1.Jiang, J., Zhai, C.X.: A systematic exploration of the feature space for relation extraction. In: HLT-NAACL 2007, pp. 113–120. Association for Computational Linguistics, Rochester (2007)
  2.Zhang, J., Ouyang, Y., Li, W., Hou, Y.: A novel composite kernel approach to chinese entity relation extraction. In: Li, W., Mollá-Aliod, D. (eds.) ICCPOL 2009. LNCS, vol. 5459, pp. 236–247. Springer, Heidelberg (2009)CrossRef
  3.Plank, B., Moschitti, A.: Embedding semantic similarity in tree kernels for domain adaptation of relation extraction. ACL 1, 1498–1507 (2013)
  4.Agichtein, E., Gravano, L.: Snowball: extracting relations from large plain-text collections. In: Proceedings of 5th ACM Conference on Digital libraries, pp. 85–94. ACM, New York (2000)
  5.Li, J., Cai, Y., Wang, Q., Hu, S., Wang, T., Min, H.: Entity relation mining in large-scale data. In: Liu, A., Ishikawa, Y., Qian, T., Nutanong, S., Cheema, M.A. (eds.) DASFAA 2015 Workshops. LNCS, vol. 9052, pp. 109–121. Springer, Heidelberg (2015)CrossRef
  6.He, T., Xu, C., Li, J., et al.: Named entity relation extraction method based on seed self-expansion. Comput. Eng. 32(21), 183–184 (2006)
  7.Pantel, P., Pennacchiotti, M.: Espresso: Leveraging generic patterns for automatically harvesting semantic relations. In: Proceedings of 21st International Conference on Computational Linguistics and The 44th Annual Meeting of the Association for Computational Linguistics, pp. 113–120. Association for Computational Linguistics, Stroudsburg (2006)
  8.Guo, X.Y., He, T., et al.: Chinese named entity relation extraction based on syntactic and semantic features. J. Chin. Inf. Process. 28(6), 183–189 (2014)
  9.Li, H., Wu, X., Li, Z., et al.: A relation extraction method of Chinese named entities based on location and semantic features. Appl. Intell. 38(1), 1–15 (2013)CrossRef
  10.Nguyen, T.V.T., Moschitti, A., Riccardi, G.: Convolution kernels on constituent, dependency and sequential structures for relation extraction. In: Proceedings of 2009 Conference on Empirical Methods in Natural Language Processing, pp. 1378–1387. Association for Computational Linguistics, Stroudsburg (2009)
  11.Liu, C., Li, S.J.: Word similarity computing based on HowNet. Comput. Linguist. Chin. 7(2), 59–76 (2002)
  12.Lu, S., Bai, S., et al.: An unsupervised approach to word sense disambiguation based on sense-words in vector space model. J. Softw. 13(6), 1082–1089 (2002)
  13.Chen, C., He, L., Lin, X.: REV: extracting entity relations from world wide web. In: Proceedings of 6th International Conference on Ubiquitous Information Management and Communication, pp. 1–5. ACM, New York (2012)
  
• 作者单位：Feiyue Ye (16)
  Nan Tang (16)
  
  16. Shanghai University, Shanghai, China
  
• 丛书名：Advances in Swarm Intelligence
• ISBN：978-3-319-41009-8
• 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：9713

文摘

This paper proposes a bootstrapping-based method to extract multiple entity relations. Compared with previous entity relation extraction methods, this method analyzes the syntax and semantics of sentences based on traditional context pattern representation. In this way, the features of keyword with the nearest syntactic dependency, phrase structure distance and semantics are extracted so as to form new semantic patterns. To reduce the noise caused by pattern extension, patterns and instances are adopted to verify their reliability mutually. In addition, by combining the information entropy of patterns, accurate and significant instances are selected. Experimental results show that this method effectively improves the quality of patterns and obtains favorable extraction results.