Automatic comprehension of algorithms for algorithmic assessment
Ismail Bouacha, Tahar Bensebaa · International Journal of Innovation and Learning · 2015
The evaluation is an unpleasing load of the teacher. Usually, in programming but especially in algorithmic, many parameters must be taken into consideration (diversity of the solutions, quality standards). These parameters could make the evaluation more difficult. The promises of an automatic evaluation are really necessary to eliminate the hard task of evaluation. These promises can reduce the direct evaluation and at the same time, they can produce a common feedback of the students. Unfortunately, the automatic approaches of evaluation of the programs are not widespread. We propose an automatic evaluation approach that comprehends algorithms. This approach uses techniques of program comprehension applied in software engineering to maintain and reuse existing applications. We present the results of our first tests.