摘要
Abstract
Amidst the widespread application of big data and artificial intelligence technologies,intelligent recommendation systems are evolving from auxiliary tools into the core hub for resource scheduling in the film industry.By continuously opti-mizing precise user demand mining,dynamic resource allocation,and vertical genre selection,they contribute to the distri-bution of film content.This study deeply focuses on this domain,addressing core challenges in film transmission recom-mendations including cold start problems and data sparsity in traditional collaborative filtering algorithms,alongside content-based filtering algorithms'limitations in deeply excavating personalized user needs.Innovatively,it proposes a hierarchical fusion strategy integrating content-based and collaborative filtering approaches.This strategy incorporates data augmenta-tion and deep learning integration techniques,dynamically optimizing recommendation mechanisms to not only significantly enhance system accuracy and efficiency but also substantially optimize cloud resource utilization while reducing film con-tent transmission latency,thereby achieving optimized allocation of film resources and providing novel insights for advanc-ing intelligent recommendation systems in film distribution.关键词
自编码器/分层动态融合算法/深度学习/智能推荐Key words
Autoencoder/Hierarchical Dynamic Fusion Algorithm/Deep Learning/Intelligent Recommendation分类
计算机与自动化
