Loading...

Question Processing for Open Domain Persian Question Answering Systems

Hosseini, Hawre | 2016

609 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 48105 (31)
  4. University: Sharif University of Technology
  5. Department: Languages and Linguistics Center
  6. Advisor(s): Bahrani, Mohammad
  7. Abstract:
  8. Question answering systems are systems which get a question in natural language as input and present an explicit, appropriate answer to the question. One of the major components of automatic question answering systems is question processing component in which the input question is analyzed. The main goal of question processing phase is to determine the answer type through question classification. Rule-based, machine learning-based and hybrid approaches have been used in order to develop question classifiers among which machine learning-based ones have outperformed the others. This study’s main goal is to develop a question classifier for Persian open domain question answering systems. Meanwhile, some other sorts of information, usually extracted in question processing, are to be extracted. In order to learn the question classifier’s model, a variety of lexical, syntactic and semantic features have been exploited and SVM is used as the machine learning method. Some of the more important features include head chunk, head word and class specific related words. As a result of this study, some syntactic and semantic features were redefined for Persian language and extracted automatically. Results of the classifier have shown that among all syntactic and semantic features used in this study, class specific related words has proved the most effective feature in optimizing the classifier’s results. The best composition of features was the combination of head word and class specific related words which provided an accuracy of 85.7 on coarse classes and 74.4 on fine classes
  9. Keywords:
  10. Natural Language Processing ; Semantic Features ; Machine Learning ; Question Classification ; Syntactic Features ; Machine Learning-based Methods ; Automatic Question Answering

 Digital Object List

 Bookmark

No TOC