The Smart Grader: Grading Programming Problems for e-Learning Environments
Shehab Gamalel-Din · 2013
In the field of education, it is commonly believed that errors are valuable knowledge probes more than being failures. Therefore, many educators believe that among effective learning methods are learning-by-explaining-experience and learning-by-correcting-mistakes through one-on-one interactions with students. It is not surprising, then, that an effective way of teaching is to give students immediate feedbacks on exercises and problems that they have just solved. Unfortunately, such one-on-one teaching scenarios are becoming increasingly difficult to arrange in e-learning, especially web-based learning, where the number of participating students is continuously increasing. Computerized empowered intelligent graders who provide students with comprehensive explanations on their mistakes and what a correct answer would be are of urgent need to overcome such difficulties. Intelligent computerized Graders would analyze students’ steps in problem solving sessions and advise them when needed. This research have introduced the new concept of Intelligent Grading Systems (IGS), designed a generic framework, and implemented Smart Grader (SG) – a simple proof of concept prototype. SG is assumed to integrate to e-Learning systems to provide more effective learning through grading student tests and correcting their mistakes as well as providing advices on better ways of problem solving. SG works with Lisp programming, as an experimental domain and a proof-of-concept, though programming in other languages as well as mathematics are all possible domains. Lisp was the subject language selected because of the simplicity of its interpretation.