Scholarlink

<p>The rapid growth of online animation platforms has led to an unprecedented volume of user–content interaction data, posing new challenges for scalable modeling of sequential behaviors and real-time service intelligence. Understanding how users consume animation content over time is essential and challenging for tasks such as personalized recommendation, intelligent caching, and adaptive service orchestration in cloud environments. Considering this challenge, this paper presents CloudSeqAnim, a cloud-oriented sequential behavior modeling framework designed to support large-scale animation content services. In concrete, CloudSeqAnim integrates lightweight sequence representation, temporal dependency modeling, and cloud-based service coordination to enable efficient analysis of long interaction sequences. To validate the feasibility of our proposed CloudSeqAnim framework, we have designed a set of simulation experiments based on the well-known time-series prediction dataset, i.e., Foursquare dataset. Experimental results on the dataset demonstrate that our proposed CloudSeqAnim framework achieves competitive accuracy and favorable efficiency compared with representative sequential baselines, while maintaining scalability suitable for cloud deployment.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

CloudSeqAnim: cloud-based sequential behavior modeling for large-scale animation content services

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月18日

查看全文
  • Chao Guo,
  • Mohammad Mehdi Ghasemi

摘要

The rapid growth of online animation platforms has led to an unprecedented volume of user–content interaction data, posing new challenges for scalable modeling of sequential behaviors and real-time service intelligence. Understanding how users consume animation content over time is essential and challenging for tasks such as personalized recommendation, intelligent caching, and adaptive service orchestration in cloud environments. Considering this challenge, this paper presents CloudSeqAnim, a cloud-oriented sequential behavior modeling framework designed to support large-scale animation content services. In concrete, CloudSeqAnim integrates lightweight sequence representation, temporal dependency modeling, and cloud-based service coordination to enable efficient analysis of long interaction sequences. To validate the feasibility of our proposed CloudSeqAnim framework, we have designed a set of simulation experiments based on the well-known time-series prediction dataset, i.e., Foursquare dataset. Experimental results on the dataset demonstrate that our proposed CloudSeqAnim framework achieves competitive accuracy and favorable efficiency compared with representative sequential baselines, while maintaining scalability suitable for cloud deployment.