Application of The K-Means Clustering Method To Search For Potential Tourists of Bendesa Hotel

Abstract

Hotels play a significant role in the growth of global tourism. With intense competition in the hotel industry, hotels are shifting their focus from solely providing superior services to identifying potential tourists. In a previous study, the J48 algorithm was employed to extract hotel transaction patterns, achieving an accuracy level of 71.6418% by considering gender and age characteristics[1]. In a separate study, foreign guest ratings by province were classified into three clusters. The study concluded that nearly 90% of provinces in Indonesia exhibit low levels of tourism, supported by the analysis of the number of tourists staying, as reported by the statistical center[2]. To identify potential tourists who can bring benefits to the hotel, hotel managers can utilize the k-means algorithm. In this study, a data mining process was conducted using data collected from tourists who stayed at the Bendesa Hotel. The process began with tourist segmentation using the K-means algorithm divided into clusters. Subsequently, the accuracy of the obtained data was calculated. This research employed room class as a reference value to discover tourist characteristics at the Bendesa Hotel. The results of applying the K-means model with 4 clusters indicated that the accuracy level for identifying potential tourists reached 84.4%.