Please use this identifier to cite or link to this item:
Title: Question classification and retrieval for english grammatical question
Authors: Tan, Jackson Wei Li.
Keywords: DRNTU::Engineering::Computer science and engineering
Issue Date: 2013
Abstract: The purpose of this project is to improve grammar categorization system that categorize English fill-in-the blank questions into different categories in order to provide a more accurate information retrieval system for users to query relevant questions from the determined categories in the database. The system targets on questions from National Higher Education Entrance Examination, Gaokao (高考), which is an academic examination held annually in China. This examination is a prerequisite for entrance into almost all higher education institutions at the undergraduate level. The system uses MorphAdorner part-of-speech tagger, a Java API that uses rule-based approach to identify English sentence patterns. This project basically split into two parts. One of them is to focus on automatic classification of question based on the sentence pattern, which assigns every question in the database to a specific category. The other one is to focus on the retrieval of the relevant questions based on user query. An experiment is conducted for both the question classification and question retrieval to evaluate the overall performance. The system is also built with an interface to show the overall performance of the question classification. An analysis is done at the end of the experiment to give a short discussion on the performance. Finally, a short conclusion is given at the end of this report as well as some recommendations for further improvement of the system in the future.
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.91 MBAdobe PDFView/Open

Page view(s) 50

checked on Sep 23, 2020

Download(s) 50

checked on Sep 23, 2020

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.