A Hotel Recommender System Based on Multi-Criteria Collaborative Filtering

Abstract

Recommendation systems have lately gained popularity in a variety of applications due to their ability to operate as information filters, thus delivering useful suggestions to users based on the processing of a variety of information from various sources. The tourism industry, on the other hand, is becoming increasingly popular, with significant growth in the usage of online services for hotel selection and reservation. Potential travelers may, however, find that using such online services is inconvenient and time-consuming. This paper aims to develop a novel fusion-based multi-criteria collaborative filtering model that provides more effective and personalized hotel recommendations. The proposed model enhances the prediction accuracy of hotel recommendations by the deployment of multi-criteria ratings that precisely express travelers’ complex preferences and addresses the insufficiency of rating information in the hotel domain by the exploitation of the users’ and items’ implicit similarity, users’ similarity propagation, and user/item reputation concepts. The experimental results demonstrate that the proposed model provides higher recommendation accuracy and coverage compared to other benchmark recommendation algorithms.