Abstract: In the realm of education, the creation of question papers is a fundamental yet time-consuming task for educators. With the advancements in Natural Language Processing (NLP), automated systems can now assist in generating question papers efficiently and effectively. This paper proposes a Question Paper Generator (QPG) that utilizes NLP techniques to analyze and generate questions tailored to specific educational domains. The QPG employs various NLP tasks such as text summarization, keyword extraction, and semantic analysis to understand the content of educational materials. By processing textbooks, lecture notes, and other relevant resources, the system identifies key concepts and formulates questions that assess the students' understanding of the subject matter. Additionally, the QPG ensures the questions adhere to the prescribed curriculum and learning objectives. Furthermore, the QPG incorporates features for customization, allowing educators to specify parameters such as question types (e.g., multiple choice, short answer, essay), difficulty levels, and topic preferences. Through this flexibility, the system can generate question papers that meet the diverse needs of different educational settings. Evaluation of the QPG involves comparing the generated question papers with those created manually by subject matter experts. Metrics such as accuracy, diversity, and relevance of questions are assessed to validate the effectiveness of the system. Additionally, user feedback from educators and students is gathered to refine and improve the QPG over time. The Questions are generated into two types such as subjective and objective. The application is executed using the framework called Flask. Overall, the proposed Question Paper Generator leveraging NLP techniques presents a promising solution to streamline the question paper creation process, thereby saving educators' time and ensuring the quality and relevance of assessment materials in educational settings. home.

Keywords: Question generator, NLTK, Natural Language Processing, POS Tagging, Flask.

| DOI: 10.17148/IARJSET.2024.11519