Analysis of customer purchasing behavior in E-commerce model

Because of globalization and information technology (IT), e-commerce (EC) has changed business practices and had an impact on established market systems. Global EC sales are still increasing, which emphasizes how crucial it is to comprehend consumer purchasing behavior. Using a Kaggle dataset named "Exploring E-commerce Trends," which contains product data including price, rating, stock amount, and sales, this paper examines consumer purchasing behavior. The paper used Multiple Linear Regression and the Random Forest methods. The quantity of stock, price, and the number of reviews were found to be important predictors of sales using the Random Forest ensemble learning technique. However, the model exhibited overfitting, performing well on the training set but poorly on the testing set. Multiple Linear Regression results showed minimal explanatory power for sales, indicating limitations in the model's effectiveness. While the Random Forest model identified key factors influencing sales, its overfitting and the ineffectiveness of the Multiple Linear Regression model suggest the need for more robust methods to predict customer purchasing behavior in e-commerce accurately. Future research should integrate additional variables and employ more sophisticated models to enhance prediction accuracy and business decision-making.