Towards an adaptive and personalized assessment model based on ontologies, context and collaborative filtering

Abstract

The assessment phase plays a very important role during the students’ teaching-learning processes since from this phase the knowledge acquired by them are validated and the shortcomings and/or strengths from students are detected. However, to do so, the questions selection made by the teacher or the learning platform does not always respond to the needs, limitations and/or cognitive characteristics of the students. In this context, it become necessary the incorporation of mechanisms that allows to obtain the main student features in a better way in order to use them during the process of question selection. In fact, this brings several benefits such as a better acquired knowledge measurement, an increase in the students’ interests, a better fail detection for new educational resource recommendation, among others. In order to make a better question selection that fulfil the student’s needs, this paper aim at proposing a characterization of the most relevant techniques and models for question selection. Likewise, an ontological model of personalized adaptive assessment is proposed, supported by Artificial Intelligence techniques that incorporate relevant cognitive and contextual information of the student to carry out a better selection and classification of questions during the e-assessment process.