Social Network and Sentiment Analysis for Enhancing Social CRM in Indonesian Educational Technology Platforms

The rapid advancement of digital technology has significantly transformed the education sector, including in Indonesia. According to the 2024 report by Badan Pusat Statistik (BPS), e-learning is among the primary reasons Indonesians access the internet. This trend has positioned educational technology (EdTech) platforms such as Ruangguru, Pahamify, and Zenius as key players in the country’s e-learning ecosystem. Simultaneously, social media has become a space where users actively express their experiences regarding the services they use. This study aims to examine user interaction dynamics and public sentiment toward these three EdTech platforms through an integrated approach combining Social Network Analysis (SNA) and Lexicon-Based Sentiment Analysis. Data were collected from platform X and preprocessed for analysis. Network analysis used Gephi to evaluate structural properties and centrality measures, while sentiment analysis used a combination of the InSet lexicon and user-generated vocabulary. To further capture discussion themes, topic modeling using the BERTopic algorithm was also applied to categorize dominant topics from user conversations. The results show that each platform exhibits different social network characteristics. Zenius demonstrates efficient information flow, Ruangguru displays tightly connected user interactions, and Pahamify presents a more dispersed structure. Overall, the sentiment analysis showed that Ruangguru and Zenius had relatively higher proportions of positive sentiment, with 44.6% and 41.4%, respectively. These findings highlight how integrating SNA and sentiment analysis can form a strong foundation for developing Social CRM strategies to enhance the quality of digital education services in Indonesia.