README.md

metadata

title: Amazon Reviews Sentiment Polarity
emoji: 🛍️
colorFrom: pink
colorTo: red
sdk: docker
app_file: app.py
pinned: false

🛍️ Amazon Reviews Sentiment Polarity App

This is a Streamlit app that performs sentiment analysis on Amazon product reviews using a fine-tuned BERT model.

🔗 Live Demo: View on Hugging Face Spaces

---

🚀 Features

• 🔎 Global Sentiment Analysis — Predicts the overall sentiment of the entire review.
• ✂️ Sentence-Level Predictions — Splits the review into sentences and classifies each individually.
• 🎨 Color-Coded Output:
  • ✅ Positive → green highlight
  • ❌ Negative → red highlight
  • ⚪ Neutral (score < 0.6) → no highlight
• 🖱️ Hover Animation — Hover over any sentence to see the sentiment confidence score.

---

📚 Model Used

• AdamCodd/distilbert-base-uncased-finetuned-sentiment-amazon
  • Hosted on Hugging Face 🤗
  • Fine-tuned for sentiment analysis on Amazon reviews

---

🧪 Example Review

"The build quality is decent. I absolutely love the screen! The battery is okay. But the software is very buggy."

• Global Prediction: Negative
• Shows sentence-by-sentence classification with hover scores and color highlights.

---

🧰 How to Run Locally

git clone https://huggingface.co/spaces/GGOSinonD/Text-Mining-Amazon-Reviews-Polarity
cd Text-Mining-Amazon-Reviews-Polarity
pip install -r requirements.txt
streamlit run app.py