Question Crafting System for Personalized Learning using Large Language Model

It is a time- and effort-consuming task for the educators to manually craft assessment questions. This project suggests a Question Crafting System for Personalized Learning using large language models (LLMs). The project aims to automatically generate personalized questions from user-uploaded content (PDF, DOCX, TXT). It efficiently extracts the most important information. from the content using text summarization, enabling personalized learning. The system uses NLP algorithms to identify the topic of the content and enable personalized questions based on the specific topic. Difficulty level, i.e., Bloom's Taxonomy levels and distractors for MCQs and question type (MCQ or True/False) can be chosen by users. The system prepares the corresponding questions accordingly. Users get a performance analysis upon solving the questions, enabling effective monitoring of learning. The system has been developed employing a Flask-based interface with an interactive and user-friendly user interface for personalized learning.