Automatic Question Generation System Using Natural Language Processing

Abstract

This project introduces an innovative approach to advance the field of automatic question generationusingnaturallanguageprocessing(NLP), withaspecificfocuson Bloom’sTaxonomy. With theincreasingavailabilityofresourcesandonlinelearningplatformsthereisaneedforefficientmethods tocreatediverseandcontextuallyrelevantquestions.Themaingoalofthisprojectistodevelopasystem that can automaticallygeneratequestionsusing Natural LanguageProcessing(NLP) techniquesaligned withfirstthreecognitivelevelsofBloom’sTaxonomy: remembering,understanding,andapplying.This project will make a contribution to the field of NLP by providing a framework for automatic question generation.Theprojectfollowsstages;preprocessingtheinputtextidentifyingconceptsandinformation creating question rules and generating different versions of questions based on these rules. This project utilizesNLPtechniquessuchasNamedEntityRecognition(NER)PartofSpeechtagging(POS),syntatic analysis and Discourse analysis. The overarching goal is to provide educators, content creators, and learners with an efficient and intelligent tool for generating questions that enhance comprehension and criticalthinking.Byautomatingthisprocess,theprojectseekstosavetimeandeffortwhileimprovingthe overall learning and assessment experience.

Introduction

Summary:

Automatic Question Generation (AQG) uses Natural Language Processing (NLP) to automatically create questions from given text, aiming to save time and improve learning and assessment experiences. AQG systems generate questions whose answers lie within the input text, supporting educational tools, content creation, and user interaction.

Literature Review:
Previous works include systems for adaptive learning with personalized question difficulty, NLP-based question generation aligned with Bloom’s Taxonomy, and studies emphasizing accurate linguistic preprocessing like POS tagging. Related AI applications in retail personalization inspire adaptive question generation approaches.

Proposed System:
The project automates question creation through two main modules:

• Customized Named Entity Recognition (NER) to extract key entities and generate questions based on Bloom’s cognitive levels.

• Rule-based templates using discourse markers and POS tagging to form why/how/who/what/where questions.

Objectives:

• Automate high-quality question generation across knowledge domains.

• Ensure question diversity and extract key information from text.

• Enhance learning and assessment efficiency by reducing manual effort.

System Architecture:

• Client-side app built with Flutter for cross-platform use.

• Backend services to process user input and manage questions.

• A PHP/XAMPP database for storing and retrieving generated questions.

• A question recognition module leveraging machine learning for accurate detection.

Implementation:

• Setup of Flutter environment and integration of question generation libraries.

• User-friendly interface with sign-up/login features.

• Input text submission for question generation following Bloom’s Taxonomy.

• Output includes diverse question formats generated from the input text.

Conclusion

In conclusion, the Automated Question Generation (AQG) project offers significant benefits by streamlining the process of creating educational assessments, quizzes, and practice materials. It saves time for educators, enhances personalized learning,and provides instant feedback to learners. While AQG systems are highly effective in generating diverse question types quickly, they still require refinement to improve creativity and contextual understanding. By integrating advanced natural language processing techniques and leveraging user feedback, AQGsystems can continue to evolve, playing a vital role in modern education, training, and assessment environments.

References

[1] DannyC.L.Tsai,AnnaY.Q.Huang, OwenH.T.LuandStephenJ.H.Yang, ”Automatic Question Generation for Repeated Testing to Improve StudentLearningOutcome”,In2021International ConferenceonAdvancedLearningTechnologies (ICALT)IEEE DOI 10.1109/ICALT52272.2021.00108 [2] ShivaliJoshi,ParinShahandSahilShah,”Automatic Question Paper Generation, according to Bloom’s Taxonomy, by generating questions from text using Natural Language Processing”, In International Journal of Innovative Science and Research Technology— Vol.6 4, April– 2021—ISSN No:-2456-2165. [3] Deena Gnanasekaran, Raja Kothandaraman, and Kannan Kaliyan,”An Automatic Question Generation System Using Rule-Based Approach in Bloom’s Taxonomy”,InRecentAdvancesinComputerScience and Communications, 2021, 14, pp. 1477-1487. [4] U. A. Mardand, and A. Gaikwad, ”A surveyon automaticquestion paper generationsystem”,Int.Adv. Res.J.Sci.Eng.Technol.,Vol.4, no.4,pp.18-20, 2017 [5] Kalpana B. Khandale, Ajitkumar Pundage ,C. Namrata Mahender, Similarities in words Using Different Pos Taggers, IOSR Journal of Computer Engineering(IOSR-JCE)e-ISSN:2278-0661,p-ISSN: 2278-8727, PP 51-55 Published: 2019 [6] M.Blstak,andV.Rozinajova, ”Machinelearning approach to the process of question generation”, In International Conference on Text, Speech, and Dialogue, 2017, pp. 102-110.[http://dx.doi.org/10.1007/978-3-319-64206-212] [7] P. Garg, and C. S. Bedi, ”Automatic question generation system from punjabi text using hybrid approach”,Int.J.Comput.TrendsTech., Vol.21,no.3,pp.130-133,2015.[http://dx.doi.org/10.14445/22312803/IJCTT- V21P125] [8] Ghuge Archana, Rajesh Gupta, Mohammed Usman, Pratibha V Kashid, Lekshmi Mohan, Vishakha A. Gaidhani , (2025). Artificial Intelligence and IoT in Retail Marketing: Innovations in Smart Stores and Personalized Shopping. Journal of Informatics Education and Research (JIER).

Copyright

Copyright © 2025 Prof. Archana R. Ghuge, Anuj Ravindra Dawkhar, Nishant Anil Bhabad, Shraddha Deepak Bodke, Ashwini Pralhad Garje. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.