Video Recommendations via NCF and DDPG Algorithms
Yu Cheng Xia, Qianzhou Chen, Zongren Liu · 2024
The rapid growth of online video content has created a need for advanced recommendation systems to boost user engagement on digital platforms. Traditional methods fall short of meeting modern users' expectations for personalized content. This paper presents an intelligent video recommendation system using neural collaborative filtering (NCF) and Deep Deterministic Policy Gradient (DDPG) algorithms. NCF utilizes neural networks to capture intricate user-item interactions, offering precise and nuanced recommendations. DDPG, a reinforcement learning technique, allows the system to adapt dynamically to users' changing preferences through real-time feedback. Empirical results demonstrate significant improvements in user engagement and retention. This comparison between NCF and DDPG highlights the transformative impact of deep and reinforcement learning on digital content curation, delivering highly personalized and engaging user experiences.