Man vs machine: The efficacy of experience-based pricing strategies in hotel revenue management using active learning with machine learning

This study proposes a machine learning (ML) framework enhanced with an active learning (AL) algorithm to replicate the decision-making processes of experienced revenue managers and improve hotel price forecasting. Using real sales data from a five-star hotel in Malaysia, four ML models are integrated with AL to enhance pricing accuracy. Model performance is assessed through revenue validation and feasibility analysis. To interpret model predictions and identify key drivers of pricing decisions, this study employs Pearson correlation analysis and SHAP analysis. Results indicate that the AL-enhanced ML model effectively learns and simulates revenue managers’ pricing adjustment behaviors, reduces prediction bias, and increases revenue. This research advances AI-driven dynamic pricing in hotel revenue management and offers practical insights for academia and industry.