Mobile-based Recommendation System for the Tour Package Using the Hybrid Method
DOI:
https://doi.org/10.3991/ijim.v12i8.9483Keywords:
Tour package, Recommendation system, Case Based Reasoning, Dempster Shafer Theory, Naïve Bayes, Bayes Theorem, Web Responsive, k-Fold Cross ValidationAbstract
Although it has been a lot of research recommendation of the tourist attraction, there has been no research that discusses the recommendations of tour packages from a collection of travel in the past. Therefore, in this study it is important to conduct a related study 1) The development of a mobile recommendation system using the Hybrid Method. 2) Test system accuracy in providing tour package recommendations.
The study is using CBR stages in providing travel package recommendations from a collection of travel in the past. There are 4 stages of the process: Retrieve, Reuse, Revise, and Retain. In this study the main focus on the retrieve stage using the method hybrid method. The hybrid method of the mobile recommendation system is the combination of the Naive Bayes method, Bayes Theorem, and Dempster Shafer. Where Naive Bayes is used for calculating the probability of continuous criteria such as age and frequency of visits. The Bayes theorem is used for calculating the probability such as country, gender, and visiting purpose. To determine the mass value of the combination of evidence using the Dempster Shafer method. Based on system accuracy test, stated that the total system accuracy in giving recommendation is 95% consisting of 2 kinds of accuracy is 46% full accuracy and 49% of half accuracy. While the error rate of the system in providing tour package of 5%.
