Optimizing E-Commerce Pricing Strategies: A Comparative Analysis of Machine Learning Models for Predicting Customer Satisfaction

Optimizing pricing strategies in e-commerce through machine learning is crucial for enhancing customer satisfaction and achieving business success. This study evaluates the effectiveness of five machine learning models—Linear Regression, Decision Trees, Random Forest, Support Vector Machines (SVM), and Neural Networks—in refining e-commerce pricing strategies using a dataset of historical transaction records. Models were assessed based on Mean Absolute Error (MAE), Root Mean Square Error (RMSE), R-squared (R²), and F1-Score.Neural Networks demonstrated superior performance with the lowest MAE (0.126), RMSE (0.155), and the highest R² (0.84) and F1-Score (0.88), highlighting its capacity to model complex, non-linear relationships. However, its high computational demands may limit its feasibility for some businesses. In contrast, Random Forest, with an MAE of 0.130, RMSE of 0.160, R² of 0.82, and F1-Score of 0.86, offers a balanced alternative, combining strong performance with greater interpretability. The findings emphasize the importance of choosing a machine learning model that aligns with business needs, resource constraints, and the trade-off between accuracy and interpretability. Integrating these models can optimize pricing strategies, better meet customer expectations, and improve business outcomes.