The Integration of Syntactic Parsing and Semantic Role Labeling

Proceedings of the 9th Conference on Computational Natural Language Learning (CoNLL), pages 237–240, Ann Arbor, June 2005. c⃝2005 Association for Computational Linguistics The Integration of Syntactic Parsing and Semantic Role Labeling Szu-ting Yi University of Pennsylvania 3330 Walnut Street Philadelphia, PA 19104 USA szuting@linc.cis.upenn.edu Martha Palmer University of Pennsylvania 3330 Walnut Street Philadelphia, PA 19104 USA mpalmer@linc.cis.upenn.edu Abstract This paper describes a system for the CoNLL-2005 Shared Task on Semantic Role Labeling. We trained two parsers with the training corpus in which the se- mantic argument information is attached to the constituent labels, we then used the resulting parse trees as the input of the pipelined SRL system. We present our re- sults of combining the output of various SRL systems using different parsers. 1 Introduction Semantic parsing, identifying and classifying the se- mantic entities in context and the relations between them, potentially has great impact on its downstream applications, such as text summarization, question answering, and machine translation. As a result, se- mantic parsing could be an important intermediate step for natural language comprehension. In this pa- per, we investigate the task of Semantic Role Label- ing (SRL): Given a verb in a sentence, the goal is to locate the constituents which are arguments of the verb, and assign them appropriate semantic roles, such as, Agent, Patient, and Theme. Previous SRL systems have explored the effects of using different lexical features, and experimented on different machine learning algorithms. (Gildea and Palmer, 2002; Pradhan et al., 2005; Punyakanok et al., 2004) However, these SRL systems generally extract features from sentences processed by a syn- tactic parser or other shallow parsing components, such as a chunker and a clause identifier. As a result, the performance of the SRL systems relies heavily on those syntax-analysis tools. In order to improve the fundamental performance of an SRL system, we trained parsers with training data containing not only syntactic constituent infor- mation but also semantic argument information. The new parsers generate more correct constituents than that trained on pure syntactic information. Because the new parser generate different constituents than a pure syntactic parser, we also explore the possibility of combining the output of several parsers with the help of a voting post-processing component. This paper is organized as follows: Section 2 demonstrates the components of our SRL system. We elaborate the importance of training a new parser and outline our approach in Section 3 and Section 4. Finally, Section 5 reports and discusses the results. 2 Semantic Role Labeling: the Architecture Our SRL system has 5 phases: Parsing, Pruning, Ar- gument Identification, Argument Classification, and Post Processing. The Argument Identification and Classification components are trained with Sec 02- 21 of the Penn Treebank corpus. 2.1 Parsing Previous SRL systems usually use a pure syntactic parser, such as (Charniak, 2000; Collins, 1999), to retrieve possible constituents. Once the boundary of a constituent is defined, there is no way to change it in later phases. Therefore the quality of the syn- tactic parser has a major …

Content truncated to 500 words