This paper introduces a personalized hotel recommendation system that leverages machine learning algorithms to capture and predict individual user preferences. The proposed framework generates highly customized recommendations by integrating user ratings, hotel attributes, and contextual information. To facilitate effective grouping of similar hotels, the system employs clustering techniques such as k-means, capitalizing on unlabeled data to uncover intrinsic patterns. This unsupervised approach offers flexibility and scalability compared to traditional supervised methods, which often rely on extensive labeled datasets. Experimental results confirm the system’s improved accuracy and relevance in generating hotel recommendations, highlighting its potential to address limitations found in existing solutions and to offer a comprehensive, user-centric booking experience. |
*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.