🤖 AI Summary
To address the “paradox of choice” induced by content overload in video-on-demand platforms, this paper proposes an implicit-feedback-based graph neural recommendation method. We construct a user–video bipartite interaction graph where edge weights reflect behavioral signals—such as viewing percentage—and integrate degree, closeness, and betweenness centrality to quantify node importance. Crucially, we introduce the first integration of Louvain modularity clustering with a user-centric graph ranking mechanism, enabling personalized recommendations without reliance on explicit ratings. This approach overcomes the fundamental limitation of conventional collaborative filtering methods that depend on explicit feedback, thereby substantially improving recommendation accuracy. Empirical evaluation on a documentary streaming platform demonstrates significant performance gains: click-through rate increases by 63%, completion rate by 24%, and user satisfaction by 17%, all outperforming baseline models—including Naïve Bayes and SVM—by substantial margins.
📝 Abstract
The proliferation of video-on-demand (VOD) services has led to a paradox of choice, overwhelming users with vast content libraries and revealing limitations in current recommender systems. This research introduces a novel approach by combining implicit user data, such as viewing percentages, with social network analysis to enhance personalization in VOD platforms. The methodology constructs user-item interaction graphs based on viewing patterns and applies centrality measures (degree, closeness, and betweenness) to identify important videos. Modularity-based clustering groups related content, enabling personalized recommendations. The system was evaluated on a documentary-focused VOD platform with 328 users over four months. Results showed significant improvements: a 63% increase in click-through rate (CTR), a 24% increase in view completion rate, and a 17% improvement in user satisfaction. The approach outperformed traditional methods like Naive Bayes and SVM. Future research should explore advanced techniques, such as matrix factorization models, graph neural networks, and hybrid approaches combining content-based and collaborative filtering. Additionally, incorporating temporal models and addressing scalability challenges for large-scale platforms are essential next steps. This study contributes to the state of the art by introducing modularity-based clustering and ego-centric ranking methods to enhance personalization in video recommendations. The findings suggest that integrating network-based features and implicit feedback can significantly improve user engagement, offering a cost-effective solution for VOD platforms to enhance recommendation quality.