DEVELOPMENT OF THE ADAPTATION MECHANISM FOR THE INTELLIGENT KNOWLEDGE ASSESSMENT SYSTEM BASED ON A STUDENT’S MODEL

It is already the fact that e-learning systems which incorporate adaptation to a learner are more efficient and useful in comparison with systems which do not have adaptation features. Adaptive e-learning systems are capable of delivering study materials, sequence of knowledge units, presentation format and feedback which is most suitable for each particular student.
Typically provision of adaptation is based on the use of a student’s model that reflects specific characteristics of a learner. The paper presents the student’s model build in a concept map based intelligent knowledge assessment system (IKAS). The system has been developed by the researchers from the Department of Systems Theory and Design of Riga Technical University during last five years. It has already reached the certain level of maturity and has been used successfully in practice. Nevertheless the system has several areas where adaptation based on the student’s model could be applied in order to enhance system’s functionality and flexibility regarding a particular learner. At present those areas include selection of an initial difficulty degree of a task and presentation of the most suitable type of concept explanations for each student. Therefore, the main attention of the paper is devoted to the algorithms developed to solve the mentioned issues.
The first algorithm corresponds to the process of the selection of an initial difficulty degree of an assessment task in IKAS for each particular learner on the basis of a student’s initial knowledge level. Methods that are used to get students’ knowledge level in a specific domain are described as well.
The second algorithm selects the most suitable type of concept explanations on the basis of such student’s psychological characteristic as learning style. A learning style is defined as a characteristic of cognitive, affective, and psychological behaviour that serves as a relatively stable indicator of how a learner perceives, interacts with, and responds to a learning environment. The system acquires a personal learning style of a student by offering a modified Felder-Solomon questionnaire on learning styles and analyzing results of inquiry. A full set of mapping rules that are used to choose an appropriate type of explanations is given in the paper.
In general, the paper consists of four sections. The first section gives a brief overview about student modelling and the use of a student model. The architecture and functionality of IKAS is described in the second section. The third section presents the student model implemented in IKAS focusing on the content of the model and methods used for the collection of student’s data. Adaptation algorithms based on the student’s model are presented in details in the fourth section. At the end of the paper conclusions are given and future work is discussed.