Inspiration
Teachers often spend countless hours on the repetitive and tedious task of grading papers, leaving them with less time for meaningful student interaction and lesson planning. Simultaneously, students frequently receive generic feedback, making it difficult for them to truly understand their mistakes and improve. We were inspired to build INTELLIGRADER to bridge this gap, using AI to automate the evaluation process and provide personalized, actionable feedback, ultimately making education more efficient and effective for everyone.
Our Approach
Our approach was to create a comprehensive, end-to-end platform that serves both teachers and students. We designed a dual-user interface with separate portals for student and teacher authentication. We integrated several key technologies to achieve our goals:
- Database Design & API Connections: We built a robust backend to handle user authentication, file storage, and data retrieval, ensuring a seamless flow of information between all components.
- Document Parsing & NLP Techniques: To handle lecture summaries and paper evaluations, we developed a system that can parse various file formats (PPTs, PDFs) and use natural language processing (NLP) to understand the content and questions.
- Automation Pipeline: We created an automated workflow that, once a teacher uploads a paper and answer key, evaluates the student's response, assigns a score, and generates specific suggestions.
Features & How They Are Useful
Our platform is divided into two main sections, each with unique features designed to solve specific problems:
Student Side
- Lecture Summariser: Students can upload a PDF or PPT file, and our system instantly generates a concise, easy-to-read summary.
- Usefulness: This saves students significant time during exam preparation and quick revisions, allowing them to focus on key concepts.
- Content Chat: This feature allows students to chat directly with the content of their uploaded documents.
- Usefulness: It provides a highly focused and relevant learning experience, as the chatbot's answers are limited to the uploaded material, helping students clarify doubts efficiently.
- Result & Improvements: After a teacher evaluates their paper, students can view their score and receive personalized suggestions for improvement.
- Usefulness: This provides more than just a grade; it gives students actionable insights into their mistakes. We also provide personalized YouTube video suggestions directly related to their incorrect answers, offering immediate resources for targeted learning.
Teacher Side
- Student Evaluator: This is the core of our platform. Teachers can upload a student's answer sheet and an optional answer key. Our system automatically grades the paper, assigns marks, and generates detailed feedback. Usefulness:This feature drastically reduces the time and effort teachers spend on grading, allowing them to redirect their focus toward teaching and personal student mentorship. The automatic feedback generation ensures consistency and provides a clear breakdown of each student's performance.
Built With
- css
- natural-language-processing
- openi
- postgresql
- streamlit
- youtubeapi
Log in or sign up for Devpost to join the conversation.