This project is a complete Exploratory Data Analysis (EDA) and Visualization project based on the Zomato Bangalore Restaurant Dataset.
As a 2nd-year B.Tech CSIT student learning Data Science, I built this project to improve my skills in:
- Python Programming
- Data Cleaning
- Exploratory Data Analysis (EDA)
- Data Visualization
- Business Insights & Storytelling
The project focuses on analyzing restaurant ratings, cuisines, pricing trends, online ordering, and area-wise restaurant distribution using real-world Zomato data.
- Practice complete EDA on a real-world dataset
- Improve data cleaning and preprocessing skills
- Discover trends in ratings, cuisines, and pricing
- Analyze the impact of online ordering and table booking
- Build strong visualization and storytelling skills
- Create a portfolio-ready Data Science project
The dataset contains:
- Restaurant Name & Type
- Ratings (Out of 5)
- Number of Ratings
- Average Cost for Two People
- Online Order Availability
- Table Booking Availability
- Cuisine Types
- Area & Address Information
- Rows: ~7,100
- Columns: 12
- Python
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Jupyter Notebook
Restaurant-Analysis/
│
├── data/
│ └── zomato.csv
│
├── notebooks/
│ └── Restaurant_Analysis.ipynb
│
├── images/
│
├── README.md
├── requirements.txt
└── .gitignore- Removed unnecessary columns
- Handled missing values
- Fixed data types
- Removed duplicate records
- Standardized categorical data
- Ratings distribution
- Highest rated restaurants
- Average ratings by restaurant type
- Most common restaurant types
- Restaurant type vs ratings
- Cost distribution
- Cost vs ratings relationship
- Budget vs premium restaurants
- Online ordering trends
- Table booking availability
- Impact on restaurant ratings
- Top restaurant areas
- Most expensive dining areas
- Highest rated areas
- Most popular cuisines
- Highest rated cuisines
- Heatmap for numeric features
- Relationship between ratings, cost, and reviews
- Most restaurants have ratings between 3.5 and 4.2
- Online ordering restaurants generally receive more ratings
- North Indian and Chinese cuisines are the most common
- BTM, Koramangala, and HSR Layout have high restaurant density
- Higher cost does not always mean better ratings
- Table booking is available in only a small percentage of restaurants
Through this project, I learned:
- Working with real-world messy datasets
- Data cleaning and preprocessing techniques
- Handling missing values and duplicates
- Creating visualizations using Matplotlib and Seaborn
- Extracting meaningful business insights from data
- Build an interactive dashboard using Streamlit
- Add Plotly interactive visualizations
- Perform recommendation analysis
- Apply Machine Learning models for rating prediction
git clone https://github.com/suyash-codez/Restaurant-Analysis.git
cd Restaurant-Analysispip install -r requirements.txtPlace zomato.csv inside the data/ folder.
Open notebooks/Restaurant_Analysis.ipynb using Jupyter Notebook or VS Code.
Hi, I'm Suyash Verma — a 2nd-year B.Tech CSIT student passionate about:
- Data Science
- Python
- Machine Learning
- Data Visualization
I enjoy building real-world analytics projects using Python and real datasets.
- GitHub: suyash-codez
- LinkedIn: Suyash Verma
⭐ If you found this project useful, feel free to star the repository!