Zomato Bangalore Restaurants Case Study

Overview

This dataset provides information about restaurants in Bangalore listed on Zomato. The data includes various categories of restaurants such as Buffets, Cafes, Delivery services, Desserts, Dine-out options, Drinks & nightlife, and Pubs & bars. It aims to help users identify the best restaurants based on their preferences and requirements.

About Zomato

Zomato is an Indian multinational restaurant aggregator and food delivery company, founded by Deepinder Goyal and Pankaj Chaddah in 2008. The platform provides:

• Information about restaurants.
• Menus and user reviews.
• Food delivery options from partner restaurants in selected cities.

Dataset Information

The dataset focuses on restaurants in Bangalore working with Zomato and was collected in two phases via web scraping.

Data Columns

The dataset includes the following columns:

Column Name	Description
url	The URL of the restaurant's Zomato page.
address	The address of the restaurant.
name	Name of the restaurant.
online_order	Indicates if online ordering is available (Yes/No).
book_table	Indicates if table booking is available (Yes/No).
rate	User rating of the restaurant.
votes	Total number of votes received for the restaurant.
phone	Contact number of the restaurant.
location	Neighborhood or area where the restaurant is located.
rest_type	Type of restaurant (e.g., Casual Dining, Café, Pub).
dish_liked	Most liked dishes mentioned by users.
cuisines	Types of cuisines offered by the restaurant.
approx_cost(for two people)	Approximate cost for two people.
reviews_list	List of reviews for the restaurant.
menu_item	List of menu items available.
listed_in(type)	Category of the restaurant (e.g., Buffet, Café, Delivery).
listed_in(city)	City or locality in Bangalore where the restaurant is listed.

Data Cleaning Process

To ensure the dataset is clean, consistent, and ready for analysis, the following steps were undertaken:

1. Deleting Redundant Columns:
   Removed unnecessary columns that do not provide significant value for analysis or are irrelevant.

2. Renaming Columns:
   Updated column names to make them more descriptive and easier to understand.

3. Dropping Duplicates:
   Identified and removed duplicate rows to ensure data integrity.

4. Cleaning Individual Columns:

   • Standardized text data (e.g. converting to lowercase).
   • Corrected spelling errors in categorical values.

5. Handling Missing Values:

   • Removed rows with NaN values to maintain consistency.
   • Where appropriate, fill missing values with default or calculated values.

6. Additional Transformations:

   • Converted data types (e.g., from string to numeric) for analysis compatibility.
   • Extracted useful information from combined columns.