E-Commerce-data-Linear-Regression-

This project explores the relationship between various features of an e-commerce website and the yearly amount spent by customers. The main objective is to predict the yearly amount spent by customers based on different factors.

Dataset

The dataset used in this project is named "Ecommerce Customers" and is loaded using the Pandas library.

Dependencies

• pandas
• numpy
• seaborn
• matplotlib
• scikit-learn

Project Structure

• Ecommerce Customers: CSV file containing the dataset.
• Ecommerce_Linear_Regression.ipynb: Jupyter Notebook file with the code for data analysis and linear regression modeling.

Analysis and Modeling

• Data is explored using various visualizations, including joint plots.
• Features such as 'Avg. Session Length,' 'Time on App,' 'Time on Website,' and 'Length of Membership' are used to predict 'Yearly Amount Spent.'
• The dataset is split into training and testing sets using train_test_split.
• Linear Regression model is implemented using scikit-learn's LinearRegression.
• The coefficients of the model are examined, and predictions are made on the test data.

Results

A scatter plot is created to visualize the actual vs. predicted values on the test data.

# Scatter plot
import matplotlib.pyplot as plt

plt.scatter(x=y_test, y=predicted_val)
plt.xlabel('Actual Yearly Amount Spent')
plt.ylabel('Predicted Yearly Amount Spent')
plt.title('Actual vs Predicted Yearly Amount Spent')
plt.savefig('scatter_plot.png')  # Save the plot as an image
plt.show()