Review of Automatic Computer Program Evaluation and Assessment Tools and Methods

Abstract

Understanding the computer language is more important in today’s world. All Computer Science students must have practical and proficient programming skills, which can be obtained through intensive exercise practices. Due to the regular rise in the number in a class, the evaluation of programming exercises imposes a heavy toll on the teacher or instructor, mainly if it must be performed manually. Manual grading for programming assignments might be time-consuming and error-inclined. Already available tools generate remarks with failing test instances. This research includes a thorough literature on the evolution of the recent (2004–2022) development of automatic programming assignment grading systems. From both a pedagogical and a technical viewpoint, the primary aspects supported by the tools and their diverse techniques were examined.  In conclusion, several new systems are being built while also acknowledging the underlying causes of this situation. Building open-source systems and collaborating on their expansion is recommended as one solution. This paper concludes with suggestions for future research paths and possible enhancements to automatic code evaluation.