An interactive Streamlit web app that analyzes Zomato restaurant data to provide insights into pricing, ratings, services, and customer patterns.
The app uses clustering, association rules, sentiment analysis (optional), and advanced visualizations.
✅ Filters:
- Restaurant type (Delivery / Dine-in)
- Clusters (KMeans groups)
- Price range slider
✅ Visualizations:
- Bar chart: Top 10 most common restaurant names
- Pie chart: Delivery vs Dine-in share
- Heatmap: Average rating by price band and type
- Bubble chart: Cost vs rating with vote size
- Box plot: Price distribution by type
- Correlation heatmap (rating, cost, votes, sentiment if available)
- Sentiment histogram (optional — if sentiment data exists)
- Scatter plot: Cost vs rating colored by cluster
✅ Advanced Analytics:
- Clustering using KMeans
- Association rules between services
- Optional sentiment scoring on synthetic review text
✅ Built with:
- Streamlit
- Pandas
- Seaborn / Matplotlib
- scikit-learn
pip install streamlit pandas seaborn matplotlib scikit-learnstreamlit run app.py💡 Restaurants offering both online ordering and table booking tend to have higher ratings.
💡 Mid-range price bands (~₹300–₹600 for two) are associated with better customer ratings.
💡 Delivery dominates restaurant types — but adding table booking may enhance perception.
✅ Add map-based visualizations if geo-coordinates are available
✅ Deploy to Streamlit Cloud or other hosting platforms
✅ Integrate real review text for NLP sentiment analysis
- This app uses synthetic sentiment data if no real review text exists.
- The dashboard design is modular and easily extendable.
Zomato Data Analysis Dashboard built for portfolio and learning purposes.
👉 Feel free to fork, extend, or deploy!