Zomato Data Analysis

Overview

This project involves analyzing Zomato restaurant data to gain insights into restaurant performance, customer preferences, and dining trends. Using Jupyter Notebook for data analysis and Power BI for visualization, interactive dashboards were created to explore key metrics such as ratings, costs, and restaurant types.

Objectives

• Analyze restaurant ratings and customer votes to identify top-performing restaurants.
• Explore the relationship between online orders, table bookings, and restaurant popularity.
• Examine the approximate cost for two people across different restaurant types.
• Provide actionable insights for restaurant owners and customers.

Dataset

The dataset contains the following columns:

• name: Name of the restaurant.
• online_order: Whether the restaurant accepts online orders (Yes/No).
• book_table: Whether the restaurant allows table booking (Yes/No).
• rate: Restaurant rating (out of 5).
• votes: Number of customer votes.
• approx_cost(for two people): Approximate cost for two people.
• listed_in(type): Type of restaurant (e.g., Buffet, Cafes, Delivery).

Methodology

1. Data Cleaning:

   • Handled missing values and inconsistencies in the dataset.
   • Standardized the rate column to ensure consistent formatting.
   • Converted the approx_cost(for two people) column to a numerical format.

2. Exploratory Data Analysis (EDA):

   • Analyzed the distribution of ratings and votes.
   • Explored the relationship between online orders, table bookings, and restaurant popularity.
   • Examined the average cost for two people across different restaurant types.

3. Dashboard Creation:

   • Created interactive Power BI dashboards to visualize key metrics.
   • Included visualizations such as bar charts, pie charts, scatter plots, and tables.
   • Added filters for dynamic data exploration (e.g., by restaurant type, online orders, or table bookings).

Tools Used

• Python: For data cleaning and analysis in Jupyter Notebook.
• Pandas: For data manipulation and analysis.
• Matplotlib/Seaborn: For data visualization in Jupyter Notebook.
• Power BI: For creating interactive dashboards.
• Jupyter Notebook: For interactive coding and documentation.

Key Insights

1. Sum of Votes by Restaurant Type

• Dining and Buffet restaurants received the highest number of customer votes, indicating their popularity among customers.
• Cafes and Other restaurant types received comparatively fewer votes, suggesting lower customer engagement.

2. Count of Restaurant Types

• Dining and Cafes are the most common restaurant types listed on Zomato, followed by Buffet and Other categories.
• This distribution reflects the diversity of dining options available to customers.

3. Online Orders by Restaurant Type

• Restaurants that accept online orders (Yes) are more prevalent across all types, especially in Dining and Cafes.
• A smaller proportion of Buffet and Other restaurants do not accept online orders, which may limit their customer reach.

4. Restaurant Ratings

• Higher-rated restaurants (e.g., those with ratings above 4) are more common in the Dining and Buffet categories.
• Cafes and Other restaurant types tend to have lower average ratings, indicating room for improvement in customer satisfaction.

5. Approximate Cost for Two People

• Buffet and Dining restaurants tend to have higher average costs for two people, reflecting their premium offerings.
• Cafes and Other restaurant types are more affordable, making them popular choices for budget-conscious customers.

---

Files

• zomato_analysis.ipynb: Jupyter Notebook for data cleaning and analysis.
• Zomato_Dashboard.pbix: Power BI file containing the interactive dashboards.

How to Use the Dashboard

1. Download the Zomato_Dashboard.pbix file.
2. Open the file in Power BI Desktop.
3. Use the interactive filters to explore the data:
   • Filter by restaurant type, online orders, or table bookings.