Exploratory Data Analysis (EDA) of Amazon's Top 50 Best Seller Books using Python

This project performs Exploratory Data Analysis (EDA) on the Amazon Top 50 Best Seller Books dataset. The objective is to uncover trends in bestselling books by analyzing ratings, reviews, prices, genres, and publication years.

The notebook follows a complete data analysis workflow, including data cleaning, outlier treatment, visualization, and statistical analysis to generate meaningful business insights.

• Explore the Amazon Best Seller Books dataset.
• Clean and preprocess the data.
• Detect and treat outliers.
• Analyze book ratings and reviews.
• Compare Fiction vs Non-Fiction books.
• Visualize trends across different years.
• Generate insights from bestseller data.

Dataset Used

bestsellers with categories.csv

• Name
• Author
• User Rating
• Reviews
• Price
• Year
• Genre

• Python
• Jupyter Notebook
• Pandas
• NumPy
• Matplotlib
• Seaborn

The notebook includes:

• Dataset Inspection
• Missing Value Analysis
• Data Cleaning
• Statistical Summary
• Outlier Detection
• Outlier Treatment
• Data Visualization
• Genre Analysis
• Year-wise Bestseller Analysis

Outliers were identified using Box Plots for the following numerical features:

• User Rating
• Reviews
• Price

To reduce the influence of extreme values while preserving the dataset, Winsorization (Clipping) was applied by capping values between the 5th and 95th percentiles.

This approach minimizes the effect of unusually large or small values without removing observations.

The project includes:

• 📊 Box Plots
• 📈 Count Plots
• 📉 Distribution Plots
• 📌 Genre-wise Analysis
• 📅 Year-wise Bestseller Trends

• Fiction and Non-Fiction books dominate Amazon's bestseller list.
• Highly reviewed books generally receive higher user ratings.
• Most bestselling books are priced within a moderate range.
• Bestseller trends vary across publication years.
• Outlier treatment improves the reliability of statistical analysis.

AmazonTop50BooksBestSellerAnalysis/
│
├── AmazonTop50BooksBestSallerAnalysis.ipynb
├── bestsellers with categories.csv
├── README.md
└── images/ (optional)

git clone https://github.com/ritikoo7/AmazonTop50BooksBestSellerAnalysis.git

cd AmazonTop50BooksBestSellerAnalysis

pip install pandas numpy matplotlib seaborn

jupyter notebook

Open:

AmazonTop50BooksBestSallerAnalysis.ipynb

• Exploratory Data Analysis (EDA)
• Data Cleaning
• Outlier Detection & Treatment
• Statistical Analysis
• Data Visualization
• Python Programming
• Working with Real-world Datasets

• Interactive Dashboard using Plotly
• Streamlit Web Application
• Book Recommendation System
• Author Popularity Analysis
• Sales Prediction using Machine Learning

Contributions are welcome!

1. Fork the repository.
2. Create a new feature branch.
3. Commit your changes.
4. Push your branch.
5. Open a Pull Request.

After uploading the project, replace the link below with your repository URL:

https://github.com/ritikoo7/AmazonTop50BooksBestSellerAnalysis

RITIK ASHOKDAS

B.Tech – Computer Science & Engineering (Data Science & AI)

Sikkim Manipal Institute of Technology

GitHub: https://github.com/ritikoo7