Application Of Case based Recommender System In Tourism Site Selection

Abstract

Tourism is the business of providing tours and services for tourists. Each and every tourist can be specifically differentiated on various parameters in tourism site selection decision. But tourism advice in developing countries like Ethiopia has different problems such as no sufficient and knowledgeable experts to give advice to tourists. Because of this majority of tourists are not satisfied in their tour. To address such problems, this study attempts to design and develop a prototype case based recommender system for tourism site selection decision that can provide advice for domain experts and tourists to make right decision at the right time. To develop the prototype system relevant knowledge was acquired through interview from domain experts and tourists and also document analysis was employed. Then, the acquired knowledge was modeled using hierarchical structure and represented using feature value case base representation and finally implemented using jCOLIBRI programming tool. The main data source used to develop the case based recommender system for tourism site selection is previous tourist cases from ACT office. Nearest neighbor retrieval algorithm is used to measure the similarity of new case (query) with cases in the case base.Moreover, in testing and evaluating the prototype system domain experts and tourists were participated. Thus, the average performance of the prototype system through user acceptance testing is 83% and 84% by domain experts and tourists respectively. And also the performance of the prototype system is measured using recall, precision and accuracy measures and it achieves 86% of recall, 65% of precision and 87.5% of accuracy.