Skip to content

Latest commit

 

History

History
131 lines (82 loc) · 4.62 KB

readme.md

File metadata and controls

131 lines (82 loc) · 4.62 KB

Smart Diary App - Backend

This is the backend server for the Smart Diary App, a full-stack web application that allows users to write diary entries and receive personalized auto-responses based on the sentiment of their entries.

1. Backend Project Setup (Dependencies)

  • Node.js (Express) - Backend server
  • MongoDB - Database (via Mongoose)
  • JWT - Authentication
  • Axios - External API request (Google NLP, OpenAI GPT-3)
  • dotenv - Environment variables
  • bcrypt - Password hashing
  • cors - Cross-origin resource sharing
  • rate-limiter - Rate limiting for API requests (100 requests per 15 minutes)
  • express-validator - Request validation

2. MongoDB Collection Schema

  • User collection: username, email, password (hashed with bcrypt), createdAt (validation at database level + express-validator)
  • DiaryEntry collection: user_id, text_entry, auto_response, sentiment_score, sentiment_magnitude, createdAt, updatedAt

3. JWT Authentication and Authorization

  • Register/Login routes: Hash password with bcrypt. Use JWT tokens to protect routes.
  • Middleware: Check token validity and user access rights.

4. CRUD Operations for Diary Entries

  • Create: Post new diary entry
  • Read: Get all diary entries for a user
  • Update: Modify existing diary entry
  • Delete: Remove diary entry

5. Google NLP API Integration

  • Sentiment Analysis: Pass the diary entry to Google Cloud Natural Language API.
  • Response Data: Extract sentiment score and magnitude from the API response.

6. OpenAI GPT-3 API Integration (gpt-3.5-turbo engine)

  • Personalized auto-response: Based on the sentiment analysis result, generate a thoughtful and empathetic response using the OpenAI GPT-3 API.
  • The response is generated by providing a prompt that includes the text entry and sentiment information.

7. Environment Variables

  • Store sensitive information like API keys, database connection string in a .env file.

8. Data Visualization (Future Work)

  • Trend Analysis: Aggregate and return sentiment scores over time (for future use on the frontend visualization like graphs and word clouds).

Backend API Endpoints

  1. Authentication Routes (/api/auth)

    • POST /register: Create a new user, store hashed password.
    • POST /login: Validate user credentials, return JWT token.

  2. Diary Entry Routes (/api/entries)

    • GET /: Fetch all user diary entries (protected route).
    • POST /: Create a new diary entry (protected route, triggers sentiment analysis & GPT-3 response).
    • PUT /:id: Update an existing entry by ID (protected route).
    • DELETE /:id: Delete an entry by ID (protected route).

  3. Sentiment Analysis (via axios)

    • axios.post('https://language.googleapis.com/v1/documents:analyze') - Google Cloud Natural Language API.

  4. Auto-Response Generation (via OpenAI API - downloaded as a package)

    • openai.chatCompletion.create() - OpenAI GPT-3 API.

How to Download and Run the Smart Diary App Backend

Follow these steps to set up the backend on your local machine:

  1. Clone the repository: git clone https://github.com/SahandNamvar/smart-diary-app-backend.git

  2. Navigate to the backend directory: cd smart-diary-app-backend

  3. Install dependencies: npm install

  • Ensure you have Node.js and npm installed on your machine. You can check if they are installed by running: node -v and npm -v.
  • Next, install the required dependencies by running npm install.
  1. Create a .env file in the root directory and add the following environment variables: (contact me for the API keys)
MONGO_URI=<your-mongodb-connection-string>
JWT_SECRET=<your-jwt-secret-key>
GOOGLE_API_KEY=<your-google-cloud-api-key>
OPENAI_API_KEY=<your-openai-api-key>
  1. Run the backend server: npm start or npm run dev (with nodemon)
  • You should see a message saying: Server running on port 5000 🚀 and Connected to MongoDB 🟢
  1. Test the API endpoints using Postman or any other API testing tool.
  • Once the server is running, you can interact with the backend APIs via the following routes:
    • Authentication: POST /api/auth/register: Register a new user.
    • Authentication: POST /api/auth/login: Login and get a JWT token.
    • Diary Entries: GET /api/entries: Fetch all diary entries for the logged-in user.
    • Diary Entries: POST /api/entries: Create a new diary entry (automatically triggers sentiment analysis and auto-response generation).
    • Diary Entries: PUT /api/entries/:id: Update an existing diary entry by ID.
    • Diary Entries: DELETE /api/entries/:id: Delete a diary entry by ID.

Make sure to include the JWT token in the Authorization header for protected routes.