Reproduce “An Automated Multiple-Choice Question Generation Using Natural Language Processing Techniques” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Chidinma A. Nwafor, Ikechukwu E. Onyenwe

An automated system for generating multiple-choice questions from textual lesson materials was developed by Nnamdi Azikiwe University, employing TF-IDF and N-grams for keyword extraction. It successfully constructs questions with a user-friendly web interface, demonstrating high recall in identifying relevant terms for Computer-Based Testing Examinations.

Key insights

  • A hybrid approach combining statistical keyword extraction (TF-IDF, N-grams) with rule-based question formation can effectively automate MCQ generation for educational contexts.
  • Prioritizing high recall in keyword extraction, even at the cost of lower precision, is beneficial when a human-in-the-loop (teacher review) can filter irrelevant suggestions.
  • Developing a user-friendly web interface is crucial for the practical adoption and utility of automated educational tools by non-technical educators.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.