Theoretical Evaluation of Machine Learning Approaches for Hotel Recommendation

The proliferation of recommendation systems across various industries, notably in entertainment and tourism, has become increasingly prevalent. These systems serve as invaluable information filters, adeptly processing diverse data from various networks to offer consumers tailored suggestions that align with their specific requirements. In the realm of global hotel selection, travelers frequently rely on written reviews, numerical ratings, and specific areas of interest to make informed choices. Online hotel reservation platforms often employ customer-generated recommendations to rank their properties, aiming to attract a broader clientele. The primary objective of a hotel recommendation system is to accurately predict, from an array of available options, the hotel that a user is most likely to choose. To achieve this, the development of effective recommendation algorithms, such as SVM, Random Forest, and others, is crucial. This paper presents a comprehensive literature analysis, addressing challenges in online tourism resource recommendation backed by crowdsourced data, and seeks to advance the field by exploring these algorithms. Through this exploratory study, we aim to develop a model that empowers users to discover hotels aligning perfectly with their preferences. Customer feedback plays a pivotal role in this endeavour.

 Keywords: SVM, Naïve Bayes, KNN