A Hierarchical Machine Learning Model for GDS Performance Evaluation and Ranking in Hotel Distribution Systems

This paper also proposes a hierarchical machine learning (ML) model for assessing and ranking the GDSs’ performance within the hotel sector. The traditional approaches that are generally used in the evaluation of the performance include the manual method or by simple statistical models which may not be efficient in the current complex hotel distribution systems. The multi-layered model incorporates both supervised and unsupervised learning using historical data and metrics for evaluating the proposal GDS and reinforcement learning for performance adjustment in real time. The model also handles scalability and enhances the ranking precision by incorporating the multi-source data and using the anomaly detection algorithms. The application of real-time GDS performance over different types of cases indicates better results in terms of ranking accuracy and decision-making speed.