Experimental Evaluation of an Expert System for Travel Recommender Systems

Abstract

As the amount of information in the field of tourism becomes abundant, finding the desired information for decision making is increasingly difficult. Therefore, travel recommender systems became significant tools as they offer tourists a convenient opportunity to find a travel package or to make individual travel plans. Today's TRSs need to have more mathematical based and complex back-end systems, to be able to support multiple decision making styles and various customer needs. The prototype system was designed to deal with this, in a manner in which it presents products to consumers and supports non-linear decision making processes. The objective of this research is to evaluate an expert system for travel recommender systems and determine whether the system meets the expectations of potential tourists in supporting the construction of individual travel plans. The evaluation considers aspects such as multi-dimensional coverage, experience, filtering methods and the complexity of a task. The ultimate goal of the research is to realize a quantitative characterization of the performance of the proposed system in terms of its usefulness, effectiveness, efficiency and user satisfaction. This research has found that there is a great potential of the system in helping users in constructing useful travel plans which match their preferences by using a hybrid filtering approach.