🌸 KnowledgeBloom — Knowledge Retention System

A beautiful, AI-powered web app that predicts and prevents knowledge decay using spaced repetition and Claude AI.

✨ Features

• Forgetting Curve Prediction — Ebbinghaus-based retention scoring for every concept
• SM-2 Spaced Repetition — Same algorithm used by Anki, implemented in the backend
• AI Quiz Generation — BloomBot generates personalized questions for each concept using Claude
• AI Tutor (BloomBot) — Full conversational AI tutor with context about your weak areas
• AI Study Plan Generator — Personalized study schedules based on your decay data
• Smart Alerts — Proactive notifications when concepts drop below retention thresholds
• Analytics Dashboard — Charts for retention, quiz trends, and study activity
• Subject/Chapter/Concept Organization — Full hierarchical syllabus management
• Study Session Logging — Track time spent on each concept
• Streak Tracking — Daily study habit monitoring

🚀 Setup

1. Install Dependencies

pip install -r requirements.txt

2. Set Your Anthropic API Key

# Mac/Linux
export ANTHROPIC_API_KEY="your-key-here"

# Windows
set ANTHROPIC_API_KEY=your-key-here

Get your API key at: https://console.anthropic.com

3. Run the App

python app.py

Open your browser to: http://localhost:5000

4. Create an Account

Register with your email and start adding subjects, chapters, and concepts!

🏗️ Project Structure

knowledge_bloom/
├── app.py                  # Flask backend with all routes & logic
├── requirements.txt        # Python dependencies
├── instance/
│   └── knowledge_bloom.db  # SQLite database (auto-created)
├── static/
│   ├── css/main.css        # Pink girly design system
│   └── js/main.js          # Shared utilities
└── templates/
    ├── base.html           # Base layout with sidebar
    ├── landing.html        # Public landing page
    ├── auth.html           # Login/Register page
    ├── dashboard.html      # Main dashboard
    ├── subjects.html       # Subject management
    ├── quiz.html           # Quiz & review (AI-powered)
    ├── analytics.html      # Analytics & AI study plan
    ├── alerts.html         # Smart alerts
    └── ai_tutor.html       # BloomBot AI chat

🧠 How the Science Works

Ebbinghaus Forgetting Curve

Retention is calculated as: R = 100 × e^(-t/S) where:

• t = time elapsed since last study (hours)
• S = stability factor (based on SM-2 ease factor and interval)

SM-2 Algorithm

After each quiz, the system calculates:

• New ease factor (how hard the concept is for you)
• Next review interval (days until you should review again)
• Concepts rated < 3 are flagged as weak and prioritized

Alert Thresholds

• < 40% retention → Critical alert, mark as weak
• < 80% retention → Include in quiz queue
• Past next_review date → Due for review alert

🌸 Design

Pink and girly aesthetic with Playfair Display + DM Sans typography, soft shadows, and animated cherry blossom petals.