Two-Phase Semantic Role Labeling based on Support Vector Machines

Two-Phase Semantic Role Labeling based on Support Vector Machines Kyung-Mi Park and Young-Sook Hwang and Hae-Chang Rim Department of Computer Science & Engineering, Korea University 5-ka, Anam-dong, SEOUL, 136-701, KOREA {kmpark, yshwang, rim}@nlp.korea.ac.kr Abstract In this study, we try to apply SVMs to the se- mantic role labeling task, which is one of the multiclass problems. As a result, we propose a two-phase semantic role labeling model which consists of the identification phase and the clas- sification phase. We first identify semantic ar- guments, and then assign semantic roles to the identified semantic arguments. By taking the two-phase approach, we can alleviate the un- balanced class distribution problem, and select the features appropriate for each task. 1 Introduction A semantic role in a language is a semantic relation- ship between a syntactic constituent and a predicate. The shared task of CoNLL-2004 relates to recognize seman- tic roles in English (X. Carreras, 2004). Given a sentence, the task is to analyze a proposition expressed by a target verb of a sentence. Especially, for each target verb, all constituents in a sentence which fill semantic roles of the verb have to be recognized. This task is based only on partial parsing information, avoiding use of a full parser and external lexico-semantic knowledge base. According to previous results of the CoNLL shared task, the POS tagged, chunked, clause identified, and named-entity rec- ognized sentences are given as an input (Figure 1). SVM is a well-known machine learning algorithm with high generalization performance in high dimensional fea- ture spaces (H. Yamada, 2003). Also, learning with com- bination of multiple features is possible by virtue of poly- nomial kernel functions. However, since it is a binary classifier, we are often confronted with the unbalanced class distribution problem in a multiclass classification task. The larger the number of classes, the more severe the problem is. The semantic role labeling can be formu- lated as a multiclass classification problem. If we try to apply SVMs in the semantic role labeling problem, we have to find a method of resolving the unbalanced class distribution problem. Conceptually, semantic role labeling can be divided into two subtasks: the identification task which finds the boundary of semantic arguments in a given sentence, and the classificiation task which determines the seman- tic role of the argument. This provides us a hint of using SVMs with less severe unbalanced class distribution. In this paper, we present a two-phase semantic role label- ing method which consists of an identification phase and a classification phase. By taking two phase model based on SVMs, we can alleviate the unbalanced class distri- bution problem. That is, since we find only the bound- ary of an argument in the identification phase, the num- ber of classes is decreased into two (ARG, NON-ARG) or three (B-ARG, I-ARG, O). Therefore, we have to build only one or three SVM classifiers. We can alleviate the unbalanced class distribution problem by decreasing the number of negative examples, which is much larger than …

Content truncated to 500 words