Comparison of Simple Matching Coefficient and Euclidean Distance in K-Means Algorithm for Tourism Destination Classification

Abstract

The tourism sector is currently experiencing accelerated growth after being slumped due to the Covid 19 pandemic for 3 years. government assistance continues to be disbursed to restore the tourism industry. this is responded by tourism businesses by developing tourist destinations. The development includes improving existing destinations, as well as by creating new tourist destinations. in various regions, the government and the private sector are competing to build new tourist attractions, usually located near tourist destinations that are well known to the public. These tourist destinations need to be recorded by the government and tourists. in collecting data, it is necessary to make groupings based on the characteristics of these tourist destinations. To assess a destination is done using the 6A framework, namely attractions, amenities, access, availability, activities, Ancillary. The data obtained is then clustered using the K-Means algorithm, using Simple Matching coefficient and Euclidean Distance. SMC uses the principle of similarity while Euclidean uses the principle of calculating the distance between data. from the calculation results of the two methods, the resulting calculator is the same but when the Davies Bouldin Index is calculated, Euclidean Distance shows better performance.