Consumers heavily rely on social media networks for making travel decisions and actively engage with tourism brands on social media platforms. With the advancement of information technology and the emergence of social media networks, analyzing and understanding these networks and their components have become a priority for stakeholders. One of the key topics users of social media networks utilize them for is making travel decisions. Given the vast and growing number of these networks and internet websites, categorizing them from the perspective of service providers is essential for offering incentives and future collaboration. This research analyzes one of the most significant social media networks related to travel and tourism, namely TripAdvisor, using data collected from the user community. Based on the desired indicators of customers and users, clustering has been performed. The clustering process employs the Self-Organizing Map (SOM) artificial neural network in a two-stage approach with the k-means algorithm, facilitating the analysis of the resulting clusters. While various validation indices are commonly used nowadays to determine the optimal number of clusters, this study combines a multi-indicator decision-making approach and the aggregation of different indices to present an optimized model with a compensatory approach towards the indicators