<![CDATA[Minahasa Regency has great tourism potential with a variety of destinations including cultural, natural, and man-made tourism. However, tourism promotion efforts still face obstacles due to the lack of integrated media capable of grouping destination information based on tourist interests and preferences. This study aims to apply the K-Means clustering algorithm in web-based promotional media to group Minahasa tourist destinations based on the level of user interaction, which is represented by the number of likes and comments on promotional content for each destination. The research method is carried out through several stages, namely collecting tourist destination data, pre-processing interaction data, implementing the K-Means algorithm with a specified number of clusters of three according to the main categories of tourismcultural, natural, and man-made), and implementing the clustering results into a web-based evaluation system that uses the Silhouette Coefficient to evaluate the quality of cluster formation. The results show that the K-Means algorithm is able to effectively group tourist destinations into three clusters that reflect the level of popularity, making it easier for users to find destination recommendations according to their interests. Implementation in a web-based system also provides an interactive display in the form of a list of destinations per cluster and recommendations for popular destinations. Thus, this study proves that the application of K-Means can increase the effectiveness of Minahasa tourism promotion, and in the future it can be developed with the integration of real-time data from social media and comparison with other clustering algorithms.]]>
Copyrights © 2026
