CODE EVALUATION AND SUGGESTION SYSTEM

Abstract

Today, due to a large number of online courses there is a need for automatic evaluation of computer programs. Grading of computer programs is also helpful for the companies in their recruitment process where a set of programming statements are to be solved by the students.

Thus, automatic grading of open-ended responses has increasingly become a subject of research. It provides scalability and efficiency. Existing systems which are used for such assessments provide a score on the basis of number of test cases which are passed by the computer program. This does not reflect the overall abilities of the programmer. Blending the existing systems with machine learning and considering the programming styles can prove to be more useful for the evaluation of computer programs.

Thus, our system, which is capable of addressing the above-mentioned issues, has been described in the paper. The paper focuses on how our “Code Evaluation and Suggestion System” can prove to more beneficial for the purpose of automatically evaluating computer programs. It concentrates on using machine learning algorithms combined with the traditional test case system and considering the various programming styles for grading the codes. It is also capable of providing suggestions to the students in order to improve their scores.