Adaptive Test Question Selection for Web-Based Educational System

Abstract

In this paper we present a method proposed to select test questions adapting to individual needs of students in the context of Web-based educational system. It functions as a combination of three particular methods. First one is based on course structure and focuses on the selection of the most appropriate topic for learning, second uses the Item Response Theory to select k-best questions with adequate difficulty for particular learner and the last is based on usage history and prioritizes questions according to specific strategies, e.g. to filter out the questions that was recently asked. We describe how these methods evaluate user answers to gather information concerning their characteristics for more precise selection of further questions. We evaluated proposed method within our Web-based system called Flip on domain of functional programming.