Strukturalizacja wiedzy w adaptacyjnym procesie komputerowego wspomagania kontroli dydaktycznej

Computerized adaptive testing is a form of computer-based test that adapts to the student's ability level. System successively selects questions for the purpose of maximizing the precision of the exam based on what is known about the examinee from previous questions in order to determine whether the answer given was not accidental. In turn, the task of knowledge structuralization is to organize the elements of the educational program to set up a hierarchical structure. This paper proposes a combination of two solutions - the concept of applying structuralization to students knowledge adaptive testing process. The knowledge structuralization in adapting testing process has a purpose of choosing the right proportions of questions from a set of controlled topics, so that each of them was well assessed and, at the same time, taking under consideration the computed level of student achievements. Such task is very difficult, even impossible to achieve with a small and a deficient tasks bank. Using well-defined relationship between the teaching program and question bank, capabilities of the automatic modification of test tasks easiness factor, and thus excluding from the bank questions which do not contribute anything to the process of assessment, and finally using the appropriate selection and configuration of testing algorithm, teacher can achieve very high possibility to create varied adaptive tests. Presented in this paper algorithms are constantly tested and developed by the author, but even now their usage in monitoring teaching process gives very reliable results, unattainable with classic computerized testing, becoming an alternative to time-consuming oral examination.