An evolutionary software tool for evaluating students on undergraduate courses

Abstract

Evaluation methods have not changed much over time. At least with regard to so-called conventional educational institutions. Taking evaluation tests remains one of the easiest ways for implementing and offering greater security to a teacher, when it comes for evaluating and expressing his or her opinion about the state of knowledge of their students in a given subject. The classification given by a teacher should be based on objective criteria, so that there is no doubt about the merit of the student. However, this does not always happen. Automated evaluation systems, supported by intelligent tutors, provide a much more objective support for assessment processes that is independent of potential "judgments" of one or more teachers. Such feature makes them quite attractive in this educational area. In this paper, we present a computational tool - an artificial tutor - designed especially for evaluating students on undergraduate courses, using a diversified set of practical heuristics acquired during several years of lecturing on different courses. Complementarily, we also describe the way the system works, revealing the evaluation mechanisms implemented, and showing how students may interact with it during an evaluation process.