Optimizing Retail Pricing Through Random Forest Regression: A Machine Learning Approach for Dynamic Pricing

Most retailers now are implementing machine learning to achieve more effective pricing strategies. The paper objective is to find a more optimized model that can be used to predict more accurately and effectively. Random Forest Regression model is chosen because it can handle large datasets and complex interactions which makes it suitable for the task. The result obtained from experimenting the model is interpreted using SHAP (SHapley Additive exPlanations) method which shows an excellent metrics score that indicates strong predictive capability with a R2 score of 0.965 and a Mean Absolute Error of 10.038.