Loading...

Semantic Role Labeling Using Dependency Trees of Persian Sentences

Rezaei Sharifabadi, Morteza | 2014

616 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46271 (31)
  4. University: Sharif University of Technology
  5. Department: Languages and Linguistics Center
  6. Advisor(s): Khosravizadeh, Parvaneh
  7. Abstract:
  8. A semantic role labeler is a software that takes sentences as inputs and identifies the words or groups which have semantic roles such as Agent, Theme, Source, Instrument etc. The correct identification of semantic rols using computers can improve the quality of many natural language processing tasks such as information extraction, question and answering, text summarization and machine translation.Therefore a considerable amount of research has been carried out on this topic. Semantic role labelers normally use features extracted from the syntactic structure of the input sentences. That is why the syntactic representation used has a prominent role in the system's outcome. The research carried out on semantic role labeling indicate that semantic role labelers which use full syntactic parsers always outperform systems which use shallow syntactic parsers. So far a limited amount of research has been carried out on semantic role labeling in Persian. Furthermore, the small handful of Persian semantic role labelers developed so far all use shallow syntactic parsers. In this research we produce a Persian semantic role labeler which uses a full syntactic parser based on dependency grammar and employs machine learning methods for semantic role labeling. The design of the semantic role labler provided in this research is close to state of the art systems and results show that it has a good performance
  9. Keywords:
  10. Dependency Grammar ; Machine Learning ; Natural Language Processing ; Computational Linguistics ; Semantic Role Labeling

 Digital Object List

 Bookmark

No TOC