AN NLP DRIVEN APPROACH TO OPTIMIZING MULTIPLE CHOICE QUESTION GENERATION

Abstract

The MCQ (Multiple Choice Question) Generator project aims to streamline and automate the process of generating high-quality objective questions from textual content or academic material. This project addresses the growing need for scalable and efficient methods of assessment creation in educational institutions, online learning platforms, and corporate training environments. Traditional methods of MCQ creation are time-consuming, require significant human effort, and are prone to inconsistency in difficulty and coverage. By leveraging Natural Language Processing (NLP) and machine learning techniques, this project proposes a system capable of generating grammatically correct, semantically meaningful, and pedagogically sound MCQs from input text.The essential functionality of the MCQ Generator encompasses various major components: text preprocessing, keyword extraction, question construction, distractor creation, and quality assessment. The process starts with preprocessing the input document(e.g., textbook chapter or article) in order to eliminate noise and extract meaningful information. Key terms or ideas are extracted via keyword extraction methods like TF-IDF. A paramount part of the system is producing the credible distractors-wrong options close in the context to the right option but easily identifiable. For the ease of use, the project has a user interface for the users(educators, instructors, or content developers) to enter raw text set the number of questions.