Overview of a Factual Question Generator System
Gabriela Aguilar, Cinvestav Unidad Monterrey, Mexico ; Kenji Kaijiri, Shinshu University, Japan
EdMedia + Innovate Learning, in Vienna, Austria ISBN 978-1-880094-65-5 Publisher: Association for the Advancement of Computing in Education (AACE), Waynesville, NC
The simultaneous fostering of formative assessment from the point of view of the Social Constructivism is a challenge for the teacher in science education. We assume this challenge through the implementation of the Factual Questions Generator System (FQGS). The FQGS uses text documents as its underlying knowledge source and combines various natural language processing techniques to extract and construct questions. Syntactic, semantic, and context processing will be done in order to generate questions. The implemented techniques include: a shallow parser and rules, which match the Spanish grammar patterns and it will include: named-entity recognition and co-reference resolution. The FQGS generates what, how and why questions. One interesting characteristics of this prototype is that learners will be able to decide by themselves, what are their best learning paths trough the selection of the topics that they want to evaluate. The FQGS is the result of the implementation of the first prototype of SPEBC, an adaptive computer-based assessment system.
Aguilar, G. & Kaijiri, K. (2008). Overview of a Factual Question Generator System. In J. Luca & E. Weippl (Eds.), Proceedings of ED-MEDIA 2008--World Conference on Educational Multimedia, Hypermedia & Telecommunications (pp. 5094-5099). Vienna, Austria: Association for the Advancement of Computing in Education (AACE).
© 2008 Association for the Advancement of Computing in Education (AACE)