RecallForge is a full-stack MERN application designed to help students study smarter using Active Recall and Spaced Repetition.
Users create cards for topics they learn, and the app schedules intelligent reviews to maximize memory retention and learning efficiency.
- Full-stack MERN app: MongoDB, Express, React, Node
- JWT authentication with OTP email verification and password reset via Nodemailer
- SM-2 spaced repetition algorithm for personalized study schedules
- AI-powered answer generation using Gemini API to automatically generate answers for questions logged during study sessions
- Responsive UI for quick, focused learning
- Create a Card: Enter a question and answer for each topic you learn. Each card represents knowledge you want to retain.
- AI-Assisted Learning: When logging your daily study, type in questions about what you learned. The Gemini API can automatically generate comprehensive answers, making it faster to create revision cards.
- Organize by Subject: Group cards into folders based on subjects. This allows you to filter and focus reviews on specific subjects or topics
- Intelligent Scheduling: Each card is automatically assigned a review schedule using the SM-2 algorithm.
- Daily Reviews: Visit the Review Section to see cards due for review. Attempt to recall each answer.
- Feedback & Adaptation: Rate how well you remembered each card (0–5). The system updates the next review interval for maximum retention.
- Each card tracks how well you remember it and when it was last reviewed.
- After reviewing, you rate your recall on a 0–5 scale.
- The algorithm uses this feedback to adjust the next review interval intelligently:
- Cards you remember easily appear less frequently.
- Cards you struggle with appear sooner.
- This ensures you review each card just before forgetting, maximizing retention and study efficiency.
- Fights the Forgetting Curve: Reinforces memory at optimal intervals
- Efficient Learning: Focus on material you struggle with; skip what you already know
- Active Recall: During reviews, you attempt to recall the answer from memory rather than just rereading it.
- This process strengthens memory traces and embeds knowledge deeper into long-term memory.
Backend:
- MongoDB stores cards with question, answer, EF, interval, repetition, next review date
- RESTful APIs for creating cards, fetching due reviews, updating feedback
Frontend:
- React.js for responsive, intuitive UI
- Review tab to revise all due cards one by one
- Cards tab displays all cards and allows to filter by subjects
Authentication & Security:
- JWT-based authentication
- Email verification and password reset using OTP verification
- Frontend: React.js
- Backend: Node.js, Express.js
- Database: MongoDB
- Authentication: JWT, OTP verification via Nodemailer
- AI Integration: Gemini API for intelligent answer generation