Automated Generation and Grading for Finite State Machine Assignments

Dhruva Kashyap, Manu Bhat, Ns Kumar

Abstract

With the rapidly growing student population for computer science, teachers need to be relieved from repetitive tasks like grading to focus their resources on teaching and guiding students. Plagiarism has always been a concern for instructors and has been aggravated by online environments, making grading more difficult. Our work proposes using automated tooling to address these concerns by preventing plagiarism through automating the generation of personalized, parameter-driven problem statements and grading of submissions. Our work targets a course for Theory of Computation, also called Automata theory, a core course in most Computer Science degrees. We specifically deal with finite automata-related problem statements. Our work proposes a tool to generate unique problem statements of uniform stated complexity, grade student submissions, and provide diagnostic information to students to help them improve. We address the problem statement where a student emulates a finite state machine given to them. The tool provides a scalable method to create problem statements and grade student submissions using heuristic-based test cases. This tool has been used successfully in a prestigious university.
BibTeX
@inproceedings{kashyap2022automated,
  author = {Dhruva Kashyap and Manu Bhat and NS Kumar},
  title = {{Automated Generation and Grading for Finite State Machine Assignments}},
  booktitle = {COMPUTE},
  year = {2022},
  doi = {10.1145/3561833.3561839},
}

Loading related content...

Figure