Abstract
This research urgently addresses the need to understand and manage viewer interactions with culturally significant video content, particularly the Rambu Solo ritual. By integrating the Digital Content Reviews and Analysis Framework with sentiment classification performance, toxicity score evaluation, and content analysis, the study systematically analyzed 21,562 posts across four videos, revealing critical themes related to cultural preservation and tourism impact that shaped viewer perceptions. Sentiment and toxicity evaluations of 15,762 posts showed an average toxicity score of 0.068, with a peak of 0.85174. Sentiment classification, using algorithms like SVM, k-NN, NBC, and DT, highlighted the superior performance of SVM enhanced by SMOTE, with an accuracy of 81.97%. However, the study identified limitations in automated sentiment analysis tools, noting that they may not fully capture the complexities of human expression. This research recommends incorporating advanced natural language processing techniques and multimodal analysis within the framework. This comprehensive methodology offers essential insights into the intersection of culture, tourism, and digital media, emphasizing the importance of creating and managing content that respects and promotes cultural heritage in the digital age. The findings are crucial for developing more effective strategies for digital content creation and community engagement, ensuring that cultural narratives are presented thoughtfully and respectfully to global audiences.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have