Uncovering Hidden Tourist Traits Using Characteristics-Based Clustering

The tourism sector plays a vital role in supporting the economic growth of a nation. However, the outbreak of COVID-19 slowed down businesses worldwide, with the tourism industry being one of the hardest-hit due to border closures. As the pandemic’s grip has begun to ease, there is an urgent need to revitalize the tourism sector to facilitate economic recovery. Understanding clients’ needs and preferences holds promise as a means to enhance the business. In the context of a tourism enterprise, tailoring tour packages to align with individual preferences may increase the likelihood of clients purchasing the package. In this paper, we present a machine learning-based approach to find distinct traits of tourists based on their travel behavior. The features considered include personal information and travel preferences after the COVID-19 pandemic. K-means was employed to cluster individuals into three groups, which was experimented to be the optimal number of clusters to give the highest silhouette score of 0.405. The visualization showing three distinct clusters further supports the success of the clustering. Finally, the results were analyzed to provide a detailed account of distinct characteristics of each group.