Asterios Avellas, "Recommendation system development for tourism destinations", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2020
https://doi.org/10.26233/heallink.tuc.86291
In this thesis we have the development of a multi-criteria recommendation system on hotels. The particular system takes into consideration the users’ (clients) preferences as well as the various characteristics of the hotels and, in addition evaluates and recommends it’s users’, the appropriate hotels.There has been a collection of questionnaires, in order to gather information regarding the users’ preferences. At the same time, there was a acquisition of data that had to do with the characteristics and the services of the hotels, taken directly from Booking.com. Afterwards, data cleaning has been done in order to avoid incomplete or false values.The specific data were utilized using the Agent Allocator method. We created profiles for each user and generated a multi-criteria data matrix with the assistance of a python developed a code.Furthermore, we used UTASTAR algorithm on a representative aggregate of hotels, in order to calculate the weight of each criteria. In the end, we achieved the best possible recommendation for each user, using the weight of criteria and the marginal utilities, to calculate the global utilities for every possible Hotel-User combination.