Most current approaches for automatic evaluation of source code use input/output testing to validate student-submitted solutions. However, very few use software engineering metrics to analyze source code. With the limitations of related work in mind, we present, in this paper, our 4-stage approach for automatic evaluation of source code: i) the code is first compiled and checked for any errors; ii) the compiled code is then tested against a set of JUnit tests provided by the teacher; iii) a set of software engineering metrics is used to compare the student's solution against the teacher's solution; iv) and finally, based on the previous stages, feedback is provided to the students so they can self-evaluate and identify the areas in which they need further study.

--

Computer-supported Education,Student Feedback,Automatic Evaluation,Web Application,Java Programming

• Ciências da Computação e da Informação - Ciências Naturais
• Outras Engenharias e Tecnologias - Engenharia e Tecnologia
• Engenharia Eletrotécnica, Eletrónica e Informática - Engenharia e Tecnologia