Analyzing Social Networks and Topic Clustering in Backpacker Tourism Content Reviews using K-means, Fast HDBScan, and Gaussian Mixture with Communalytic

This research explores the integration of social network analysis and topic clustering techniques to provide novel insights into digital interactions and thematic trends within the context of backpacker tourism. Utilizing a structured framework, 3,575 records across three content IDs (c2ZMFDS_3rU, Sv_yxz7T8rU, and i9t9pbdo-bk) were processed and classified into 10 clusters using k-Means, Fast HDBScan, and Gaussian Mixture algorithms. Social network analysis was performed on 4,224 actor nodes and 395 edges, highlighting the role of key influencers in driving conversations while revealing the participation patterns of a larger, less engaged audience. The topic clustering revealed distinct themes, including budget travel, off-the-beaten-path destinations, and sustainable tourism, with each algorithm offering unique insights into the structure of the data. The novelty of this research lies in applying these computational methods to backpacker tourism, traditionally analyzed through qualitative approaches, to uncover how thematic discussions propagate within digital communities. By integrating these techniques, the study provides a deeper understanding of how key topics resonate with backpackers and how social interactions influence the spread of ideas. The findings offer valuable implications for content creators and tourism marketers seeking to engage this niche travel demographic more effectively. This work contributes a scalable, data-driven methodology for analyzing traveler behavior and preferences in virtual environments, enhancing the field of backpacker tourism research.