Learning Human Language with AI

A vocabulary-learning web app that helps users build their English vocabulary through personal flashcards, dictionary lookups, AI-powered word insights, and daily language content. It combines a Django REST backend, a MongoDB database, and a vanilla HTML/CSS/JavaScript frontend, with Google Gemini providing the AI features.

---

Features

• User accounts — sign up and log in with token-based sessions; each user's flashcards are private to them.
• Flashcards — add, list, search, and delete flashcards (word, meaning, example sentence, synonyms, antonyms).
• Dictionary lookup — definitions, synonyms, and antonyms sourced from the Free Dictionary API, with Datamuse and NLTK WordNet as fallbacks.
• AI word lookup — Google Gemini generates a meaning, example sentence, synonyms, and antonyms for any word.
• Daily content — AI-generated Word of the Day, Phrase of the Day, and This Day in Language, cached per day in MongoDB.
• CLI quiz — a terminal flashcard quiz (main.py) for the MongoDB-backed deck.

---

Tech Stack

Layer	Technology
Backend	Django 5.2, django-cors-headers
Database	MongoDB (via PyMongo)
AI	Google Gemini (google-generativeai, Gemini 1.5 Flash/Pro)
NLP	NLTK (WordNet)
External	Free Dictionary API, Datamuse API
Frontend	HTML, CSS, vanilla JavaScript

---

Project Structure

.
├── main.py                     # CLI flashcard quiz entry point
├── backend/
│   ├── manage.py               # Django management entry point
│   ├── ai_agent.py             # Google Gemini integration (word/phrase/daily content)
│   ├── quiz.py                 # CLI quiz logic
│   ├── backend/                # Django project config (settings, urls, wsgi/asgi)
│   ├── api/                    # REST API app (views, urls)
│   └── database/
│       ├── db.py               # MongoDB connection & collections
│       ├── flashcards.py       # Flashcard data access
│       ├── dictionary.py       # Dictionary lookup (FreeDict → Datamuse → WordNet)
│       ├── daily_content.py    # Daily AI content with per-day caching
│       └── session_manager.py  # Token-based session handling
└── frontend/
    ├── index.html              # Vocabulary lookup page
    ├── login.html / login.js   # Auth UI
    ├── dashboard.html / .js     # Flashcard dashboard
    └── *.css, script.js        # Styles and shared scripts

---

Prerequisites

• Python 3.13 (a myenv/ virtualenv is included in the repo)
• MongoDB running locally at mongodb://localhost:27017/ (database: vocab_db)
• A Google AI (Gemini) API key

---

Setup

1. Clone and enter the project

   git clone <repo-url>
   cd "Learning Human Language with AI"

2. Create and activate a virtual environment (or use the bundled myenv/)

   python -m venv myenv
   # Windows
   myenv\Scripts\activate
   # macOS/Linux
   source myenv/bin/activate

3. Install dependencies

   pip install Django django-cors-headers pymongo google-generativeai python-dotenv nltk requests

4. Configure your API key — create a .env file in the backend/ directory:

   GOOGLE_AI_API_KEY=your_gemini_api_key_here

5. Start MongoDB (ensure the local server is running on port 27017).

---

Running the App

Backend (Django API)

cd backend
python manage.py runserver

The API is served under http://localhost:8000/api/.

Frontend

Open the HTML files in frontend/ directly in a browser, or serve them with a static file server (e.g. python -m http.server) and start from login.html.

CLI Quiz

python main.py

---

API Endpoints

All endpoints are prefixed with /api/. Flashcard endpoints require an Authorization: Token <token> header.

Method	Endpoint	Description
POST	/signup/	Register a new user
POST	/login/	Log in, returns a session token
POST	/add/	Add a flashcard
GET	/all/	List the user's flashcards
GET	/search/?query=	Search flashcards by word or meaning
DELETE	/delete/	Delete a flashcard by id
GET	/lookup-word/?word=	Dictionary lookup (FreeDict/Datamuse/WordNet)
GET	/lookup-word/ai/?word=	AI-generated word details (Gemini)
GET	/daily-content/	Daily word, phrase, and language fact

---

Notes & Caveats

⚠️ This is a learning/student project. A few things are intentionally simple and not production-ready:

• Passwords are stored in plain text (no hashing).
• MongoDB connection details and SECRET_KEY are hardcoded; DEBUG = True.
• CORS_ALLOW_ALL_ORIGINS = True.

Harden these (password hashing, environment-based secrets, restricted CORS, DEBUG = False) before any real deployment.