Predict the Success of a Restaurant on the Zomato App

Image Source: Firstpost.com

Business Understanding 😇

This project aims to gain a deeper understanding of the factors that influence Zomato's success in the food and food delivery industry. Using a dataset that records various information about the restaurants listed on the Zomato platform, we will explore trends, patterns, and key factors that can affect the popularity and performance of restaurants.

As a data scientist, my goal is to help restaurant owners, market analysts, and food industry players to understand market dynamics and analyze restaurant characteristics that can improve their success on the Zomato platform. By utilizing machine learning techniques and data analysis, we will strive to identify patterns that can be used to improve customer experience, increase sales, and optimize marketing strategies.

With a better understanding of the factors that influence the success of Zomato companies, it is hoped that restaurant owners and industry players can make more informed and effective decisions in managing their businesses. Ultimately, the project aims to make a positive contribution to the restaurant ecosystem and the food industry as a whole. Let's dive in and explore the world of Zomato! 🍽️🚀

Data Understanding 😁

This dataset is a collection of data about the restaurants listed on the Zomato platform contains in sqlite format with size 500 mb +. It includes information on various features of each restaurant, such as location, restaurant type, rating, number of votes, and more. This data can provide valuable insights into the characteristics of the restaurants and factors that might affect the success of the Zomato company.

List of features used: (Unsupervised)

• URL: The unique URL of each restaurant on the Zomato platform.
• address: The physical address of the restaurant.
• name: The name of the restaurant.
• online_order: Information on whether the restaurant accepts online orders or not.
• book_table: Information on whether or not the restaurant provides table reservations.
• rate: The average rating of the restaurant on a scale.
• votes: The number of votes the restaurant received.
• phone: The contact phone number for the restaurant.
• location: The geographical location of the restaurant.
• rest_type: The type of restaurant type (e.g. Casual Dining, Cafe, etc.).
• dish_liked: A list of dishes favored by customers.
• cuisines: The type of cuisine served by the restaurant.
• approx_cost(for two people): The approximate cost of a meal for two people at the restaurant.
• reviews_list: A list of reviews received by the restaurant.
• menu_item: A list of menu items available at the restaurant.
• listed_in(type): The type in which the restaurant is listed (e.g. Buffet, Delivery, etc.).
• listed_in(city): The city where the restaurant is listed

Data Preparation:

• Cleaning data
• Handling Missing Value
• Feature Extraction
• Feature Importance
• Univariate Analysis
• Feature Encoding
  • one hot encoding
  • mean encoding
• Feature Engineering 😢
  • Handling Outliers

Modeling 😨 😰

• Random Forest Classifier
• Decision Tree Classifier
• Logistic Regression
• KNN
• XGboost/gradient boost

Evaluation 🧐 🤓 😎

• accuracy
• recall
• f1-score
• precision