A Two-Phase Shallow Semantic Parsing System Using Clause Boundary Information and Tree Distance

절 경계와 트리 거리를 사용한 2단계 부분 의미 분석 시스템

  • Received : 2010.01.05
  • Accepted : 2010.03.19
  • Published : 2010.05.15

Abstract

In this paper, we present a two-phase shallow semantic parsing method based on a maximum entropy model. The first phase is to recognize semantic arguments, i.e., argument identification. The second phase is to assign appropriate semantic roles to the recognized arguments, i.e., argument classification. Here, the performance of the first phase is crucial for the success of the entire system, because the second phase is performed on the regions recognized at the identification stage. In order to improve performances of the argument identification, we incorporate syntactic knowledge into its pre-processing step. More precisely, boundaries of the immediate clause and the upper clauses of a predicate obtained from clause identification are utilized for reducing the search space. Further, the distance on parse trees from the parent node of a predicate to the parent node of a parse constituent is exploited. Experimental results show that incorporation of syntactic knowledge and the separation of argument identification from the entire procedure enhance performances of the shallow semantic parsing system.

본 논문은 최대 엔트로피 모형에 기반한 두 단계 부분 의미 분석 방법을 제안한다. 먼저, 의미 논항의 경계를 인식하고, 그 다음 단계에서 확인된 논항에 적절한 의미역을 할당한다. 두 단계 부분 의미 분석에서는 두 번째 단계인 논항 분류가 논항 확인 단계의 결과에 기반하여 수행되기 때문에 논항 확인의 성능이 매우 중요하다. 본 논문은 논항 확인의 성능을 향상시키기 위하여 논항 확인의 전처리 단계에 구문 지식을 통합한다. 구체적으로, 절 인식 결과로부터 술어의 인접절 및 상위절들을 확인하고, 구문 분석 결과로부터 술어의 부모 노드로부터 구문 구성 요소의 부모 노드까지의 트리 거리를 추출하여 전처리 단계에서 활용한다. 실험을 통해, 구문 지식을 활용하는 것이 부분 의미 분석 성능에 기여함과 제안하는 두 단계 방법이 한 단계 방법보다 우수한 성능을 낼 수 있음을 보인다.

Keywords

References

  1. M. Surdeanu, S. Harabagiu, J. Williams, and P. Aarseth, "Using Predicate Arguments Structures for Information Extraction," In Proceedings of the Association for Computational Linguistics, 2003.
  2. D. Gildea and D. Jurafsky, "Automatic Labeling of Semantic Roles," Computational Linguistics, vol.28, no.3, pp.1-45, 2002. https://doi.org/10.1162/089120102317341747
  3. N. Xue and M. Palmer, "Calibrating Features for Semantic Role Labeling," In Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2004.
  4. N. Kwon, M. Fleischman, and E. Hovy, "FrameNetbased Semantic Parsing Using Maximum Entropy Models," In Proceedings of the International Conference on Computational Linguistics, 2004.
  5. S. Pradhan, W. Ward, K. Hacioglu, J. Martin, and D. Jurafsky, "Semantic Role Labeling Using Different Syntactic Views," In Proceedings of the Association for Computational Linguistics, 2005.
  6. J. Gimenez and L. Marquez, "Fast and Accurate Part-of-Speech Tagging: The SVM Approach Revisited," In Proceedings of the Recent Advances in Natural Language Processing, 2003.
  7. X. Carreras and L. Marquez, "Phrase Recognition by Filtering and Ranking with Perceptrons," In Proceedings of the Recent Advances in Natural Language Processing, 2003.
  8. E. Charniak, "A Maximum-Entropy-Inspired Parser," In Proceedings of the North American Chapter of the Association for Computational Linguistics, 2000.
  9. A. Berger, S. Pietra, and V. Pietra, "A Maximum Entropy Approach to Natural Language Processing," Computational Linguistics, vol.22, no.1, pp.39-71, 1996.
  10. M. Collins, "Head-Driven Statistical Models for Natural Language Parsing," PhD Dissertation, University of Pennsylvania, 1999.
  11. S. Buchholz, "Memory-Based Grammatical Relation Finding," PhD Thesis, Tilburg University, 2002.
  12. X. Carreras and L. Marquez, "Introduction to the CoNLL-2005 Shared Task: Semantic Role Labeling," In Proceedings of the Eighth Conference on Natural Language Learning, 2005.
  13. S. Chen and R. Rosenfeld, "A Gaussian Prior for Smoothing Maximum Entropy Models," Technical Report CMUCS-99-108, Carnegie Mellon University, 1999.