🔗 Live App: https://quora-duplicate-detector-5.onrender.com/

A machine learning powered web application that detects whether two questions are duplicate / semantically similar or different. Built using Streamlit and deployed on Render.

• Detects semantic similarity between two questions
• Machine Learning based prediction
• Web interface using Streamlit
• Cloud deployment on Render
• Proper dependency management using requirements.txt

1. User enters two questions

2. Text preprocessing is applied:

   • Lowercasing
   • Stopword removal
   • Tokenization
   • Cleaning

3. Feature extraction is performed

4. Similarity distance is calculated

5. ML model predicts:

   • Duplicate ✅
   • Not Duplicate ❌

quora-duplicate-detector/
│
├── app.py              # Main Streamlit app
├── helper.py           # Helper functions
├── model.pkl           # Trained ML model
├── vectorizer.pkl      # Text vectorizer
├── requirements.txt    # Dependencies
├── README.md           # Project documentation
└── .gitignore

git clone https://github.com/your-username/quora-duplicate-detector.git
cd quora-duplicate-detector
pip install -r requirements.txt
streamlit run app.py

streamlit
scikit-learn
nltk
numpy
pandas
beautifulsoup4
bs4
distance

Deployed using Render Cloud Platform

Steps:

1. Push project to GitHub

2. Connect GitHub repo to Render

3. Select Web Service

4. Add build command:

   pip install -r requirements.txt

5. Add start command:

   streamlit run app.py

6. Deploy 🚀

Input: Q1: Machine learning helps computers learn from data.

Q2: Computers can learn from data using machine learning.

Output: ✅ Duplicate Question

This model is trained on limited data and may not be 100% accurate. Predictions should be considered as probabilistic, not absolute.

Vivek Kumar B.Tech Machine Learning Student AI/ML Developer

This project is licensed under the MIT License — free to use, modify, and distribute.

⭐ If you like this project, give it a star on GitHub