Audience behavior and engagement on instagram: A data-driven approach

This study presents a data-driven approach to understanding audience behavior and en- gagement on social media, with a focus on Instagram. By combining user survey responses with real-world post performance data collected via the Meta Graph API, the research ex- plores how users interact with content and what drives engagement. Content categories were identified using a fine-tuned BERT model, while survey data en- abled demographic segmentation and behavior pattern mining. Post insights, such as likes, views, shares, and saves, were analyzed across different time slots and content types. The findings were compared with self-reported behaviors to uncover correlations and discrepan- cies. This integrated analysis provides valuable guidance for optimizing content strategies based on audience preferences and engagement trends.