Restaurant Data Analytics Product

A beginner-friendly, end-to-end data analytics project for restaurant business insights. This repository demonstrates how to analyze a real-world restaurant dataset using Python, pandas, and visualization libraries. All code and outputs are reproducible and ready for extension or learning.

Features

• Top Cuisines Analysis: Find the most popular cuisines and their market share.
• City Analysis: Discover which cities have the most restaurants and the highest average ratings.
• Price Range Distribution: Visualize how restaurants are distributed across price categories.
• Online Delivery Impact: See how online delivery availability affects restaurant ratings.
• Votes & Popularity: Explore the relationship between customer votes and ratings.
• Service Features: Analyze how price range relates to online delivery and table booking.
• Review Text Insights: Extract frequent positive & negative keywords, review length distribution, and correlation between review length and rating.

Quick Start

1. Clone the repository
2. Install dependencies

   python -m venv .venv
   .\.venv\Scripts\activate
   pip install -r analysis\requirements.txt

3. Run the analysis

   python analysis\analysis.py

4. View results
   • Output CSVs and images are in analysis/output/
   • Open PNG files to see visualizations

Example Visualizations

• 
• 
• 
• 
• 
• 

Core Output Files (default minimal run)

These are always produced unless FULL_MODE=1 is set:

• level1_city_analysis.csv
• level1_price_range_distribution.csv
• level1_online_delivery_summary.csv
• level1_online_vs_offline_rating.csv
• level3_price_range_vs_delivery_table.csv
• level1_top_cuisines.csv
• Key charts (PNG):
  • level1_top3_cuisines.png
  • level1_city_most_restaurants.png
  • level1_price_range_distribution.png
  • level1_online_vs_offline_rating_bar.png
  • level3_votes_vs_rating.png
  • level3_price_range_vs_services.png

Additional Outputs (when FULL_MODE=1)

Top/Bottom city & cuisine lists, extended plots (top10, top15, pies, highest rating city), votes top/bottom lists, level1 summary, README_generated, review keyword extended artifacts (if provided review data).

Optional Review Analysis

Provide a reviews.csv (recommended columns: review_text, rating) in Data analysis dataset/ to enable:

• reviews_top_positive_keywords.csv / reviews_top_negative_keywords.csv
• reviews_length_stats.csv
• reviews_length_rating_corr.txt (only if rating column present)
• Associated PNG charts

Run in full mode:

set FULL_MODE=1 & python analysis\analysis.py

(PowerShell: $env:FULL_MODE='1'; python analysis\analysis.py)

How It Works

• analysis/analysis.py: Main script (supports FULL_MODE flag)
• analysis/requirements.txt: Python dependencies
• analysis/output/: Generated results (CSV + PNG)
• Data placement: put dataset CSV (any of: Dataset .csv, dataset.csv, Dataset.csv) inside Data analysis dataset/.

Extend or Learn

• Fork and add new analyses (e.g., sentiment, time trends)
• Use as a template for your own data projects
• All code is commented for easy understanding

License

MIT License. Free for learning and commercial use.