Exploring and Analyzing YouTube Communities through Data Mining and Knowledge Graphs
Purpose: The paper explores using knowledge graphs to analyze and model social interactions on the YouTube platform. The study uses advanced data structures to uncover more profound insights into community dynamics and user engagement in the digital space. Design/Methodology/Approach: The study uses a mixed-methods approach, combining real-time data extraction from YouTube's live chat feature with knowledge graph construction to map complex relationships between users, content, and interactions. The data is managed using a Neo4j graph database and processed using Redis queuing mechanisms and Kubernetes for distributed computing, providing scalability and flexibility in data handling. Findings: The study shows that knowledge graphs provide a solid framework for capturing and analyzing the complex network of social interactions on YouTube. By integrating natural language processing (NLP) techniques, the designed framework effectively processes and interprets queries and shows user interactions. Practical Implications: The study's results offer significant implications for developing more sophisticated recommendation systems and analytics tools that dynamically adapt to new data and user behavior. Implementing knowledge graphs can help platform designers and marketers better understand user engagement and content popularity, leading to more targeted and effective strategies. Originality/Value: The article contributes to the field of digital analytics by presenting a new application of knowledge graphs in social media analysis. Emphasizes the enhanced capabilities of graph-based data structures in combination with real-time data processing and NLP.