Automatic acquisition of didactic resources: generating test-based questions

Natural Language Processing (LPN) Techiniques facilitate the automatic acquisition of some didactic resources. Using corpora as source data, Arikiturri, is able to automatically generate test-based questions for Technology Supported Learning Systems. In this article we define some concepts of the generic question model underlying Arrikiturri, and explain an experiment with experts in order to evaluate the quality of the generated questions.

[1]  Montse Maritxalar,et al.  ArikIturri: An Automatic Question Generator Based on Corpora and NLP Techniques , 2006, Intelligent Tutoring Systems.

[2]  Eiichiro Sumita,et al.  Measuring Non-native Speakers’ Proficiency of English by Using a Test with Automatically-Generated Fill-in-the-Blank Questions , 2005 .

[3]  Ana Arruarte Lasa,et al.  Acquisition of the Domain Structure from Document Indexes Using Heuristic Reasoning , 2004, Intelligent Tutoring Systems.

[4]  Tom Murray,et al.  Authoring Intelligent Tutoring Systems: An analysis of the state of the art , 1999 .

[5]  Hiroshi Nakagawa,et al.  A Real-Time Multiple-Choice Question Generation For Language Testing: A Preliminary Study , 2005 .

[6]  Jason S. Chang,et al.  FAST – An Automatic Generation System for Grammar Tests , 2006, ACL.

[7]  Chao-Lin Liu,et al.  Applications of Lexical Information for Algorithmically Composing Multiple-Choice Cloze Items , 2005 .

[8]  Montse Maritxalar,et al.  The Question Model inside ArikIturri , 2007, Seventh IEEE International Conference on Advanced Learning Technologies (ICALT 2007).

[9]  Anne-Marie Vercoustre,et al.  Reusing Educational Material for Teaching and Learning: Current Approaches and Directions , 2005 .

[10]  M. Mavrikis.  MathQTI and the "Serving Mathematics" project , 2005 .

[11]  M. J. Pollock,et al.  Evaluating the costs and benefits of changing to CAA , 2000 .

[12]  Maxine Eskénazi,et al.  Automatic Question Generation for Vocabulary Assessment , 2005, HLT.