A modern React + TailwindCSS frontend for testing the Multilingual NLP Analysis API.
This project works with a backend API that fine-tunes transformer models for sentiment and toxicity detection in code-mixed Indian social media to address multilingual NLP gaps. You can find the backend repository here: https://github.com/ananikets18/code-mix-social-media-python-backend
This frontend project was bootstrapped with Create React App.
IMPORTANT: Ensure the backend API is running before starting the frontend.
# From the parent directory (NLP-project/)
python api.py
The API should be running at http://localhost:8000 (local development) or https://159-89-161-139.nip.io (production).
# Install dependencies (first time only)
npm install
# Start the development server
npm start
Open http://localhost:3000 in your browser to view the app.
- 🌐 Language Detection — Supports 60+ languages including code-mixed text like [translate:Hinglish]
- 😊 Sentiment Analysis — Positive, Negative, and Neutral sentiments with confidence scores
- 🚫 Profanity Detection — Covers 10 languages with severity levels
⚠️ Toxicity Analysis — Detects 6 categories of toxic content- 🌍 Translation — Auto-detects source language and translates between languages
- 🔤 Romanized Conversion — Converts romanized Indic text to native scripts
- 📊 Domain Detection — Identifies Technical, Financial, and Medical domains
-
📊 Analyze — Comprehensive NLP analysis
- Language detection with flags for romanized and code-mixed text
- Sentiment analysis
- Profanity detection
- Toxicity scoring
- Automatic translation
- Compact and verbose response modes
-
🌍 Translate — Translation service
- Auto-detect source language
- Select from multiple target languages
- Automatic romanized conversion support
Click any pre-filled example buttons for a quick demo:
- English — Clean English text
- [translate:Hinglish] — Code-mixed Hindi-English text
- Romanized — Romanized Indic text input
- Profanity — Test for profanity filtering
- Marathi — Native Devanagari script text
POST /analyze— Run comprehensive analysisPOST /translate— Use translation servicePOST /convert— Convert romanized text to native script
In the project directory, you can run:
Runs the app in development mode.
Open http://localhost:3000 to view it in your browser.
The page will reload if you make edits.
You will also see any lint errors in the console.
Launches the test runner in interactive watch mode.
See running tests for more info.
Builds the app for production to the build folder.
Bundles React in production mode and optimizes the build for best performance.
The build is minified, and filenames include hashes.
Your app is ready to be deployed!