文章基本信息

标题：Persian Semantic Role Labeling Based on Dependency Tree
本地全文：下载
作者：Soghra Lazemi ; Hossein Ebrahimpour Komleh ; Nasser Noroozi 等
期刊名称：INTERNATIONAL JOURNAL OF INFORMATION SCIENCE AND MANAGEMENT
印刷版ISSN：2008-8302
电子版ISSN：2008-8310
出版年度：2020
卷号：18
期号：1
页码：93-108
语种：English
出版社：REGIONAL INFORMATION CENTER FOR SCIENCE AND TECHNOLOGY
摘要：pspan style="font-family: Times New Roman;"Semantic role labeling is the task of attaching semantic tags to the words according to the occurred event in the sentence. Persian semantic role labeling is a challenging task that most methods so far in this regard depend on a huge number of handcrafted features and are done on feature engineering to attain high performance. On the other hand, by considering the Free-Word-Order and Subject-Object-Verb-Order characteristics of Persian, the verbal predicate’s arguments are often distant and create long-range dependencies. The long-range dependencies can hardly be modeled by these methods. Our goal is to achieve a better performance only with minimal feature engineering and also to capture long-range dependencies in a sentence. To these ends, in this paper a deep model for semantic role labeling is developed with the help of dependency tree for Persian. In our proposed method, for each verbal predicate, the potential arguments are identified with the help of dependency relationships, and then the dependency path for each pair of predicate and its candidate argument is embedded using the information in the dependency trees. In the next step, we employed a bi-directional recurrent neural network with long short-term memory units to transform word features into semantic role scores. Experiments have been done on the emfirst semantic role corpus in Persian language/em and the corpus provided by the authors. The achieved Macro-average Fsub1/sub-measure is 80.01 for the first corpus and 82.48 for the second one./span