EDA and Feature Engineering with Zomato Dataset

This project focuses on Exploratory Data Analysis (EDA) and Feature Engineering using the Zomato dataset to gain insights into restaurant trends, pricing, ratings, and customer preferences. Through data cleaning, visualization, and feature transformation, we prepare the dataset for predictive modeling and deeper analysis.

The dataset contains information about restaurants listed on Zomato, including:

Restaurant name

Location

Ratings

Cost for two

Cuisine types

Online delivery availability

Reviews

The following steps are performed to analyze the dataset:

Data Cleaning: Handling missing values, duplicates, and inconsistencies.

Data Visualization: Understanding trends through visualizations.

Outlier Detection: Identifying anomalies in price, ratings, etc.

Correlations: Exploring relationships between variables.

Geospatial Analysis: Studying restaurant density by location.

Key feature engineering techniques include:

Encoding categorical variables (e.g., one-hot encoding for cuisines).

Creating new features (e.g., cost-to-rating ratio, sentiment analysis on reviews).

Feature scaling (e.g., normalization for cost variables).

Handling skewness in numerical features.

Python

Pandas, NumPy

Matplotlib, Seaborn

Scikit-learn

NLP (for review sentiment analysis, if applicable)

Clone the repository:

Install dependencies:

Run the Jupyter Notebook:

Open EDA_Zomato.ipynb and follow the analysis.

Identified factors affecting restaurant ratings.

Analyzed cost distributions across different cuisines and locations.

Created new features that can enhance predictive modeling.

Contributions are welcome! Feel free to fork the repository and submit pull requests.

This project is licensed under the MIT License.