Semantics-Enhanced Temporal Graph Networks for Content Popularity Prediction

被引:4
|
作者
Zhu, Jianhang [1 ]
Li, Rongpeng [1 ]
Chen, Xianfu [2 ]
Mao, Shiwen [3 ]
Wu, Jianjun [4 ]
Zhao, Zhifeng [1 ,5 ]
机构
[1] Zhejiang Univ, Coll Informat Sci & Elect Engn, Hangzhou 310027, Peoples R China
[2] VTT Tech Res Ctr Finland, Oulu 90570, Finland
[3] Auburn Univ, Dept Elect & Comp Engn, Auburn, AL 36849 USA
[4] Huawei Technol Co Ltd, Shanghai 201206, Peoples R China
[5] Zhejiang Lab, Hangzhou 311121, Peoples R China
基金
中国国家自然科学基金;
关键词
Semantics; Predictive models; Bipartite graph; Biological system modeling; Streaming media; Computational modeling; Graph neural networks; Content caching; dynamic graph neural network; popularity prediction; semantics;
D O I
10.1109/TMC.2023.3349315
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The surging demand for high-definition video streaming services and large neural network models implies a tremendous explosion of Internet traffic. To mitigate the traffic pressure, architectures with in-network storage have been proposed to cache popular contents at devices in closer proximity to users. Correspondingly, in order to maximize caching utilization, it becomes essential to devise an effective popularity prediction method. In that regard, predicting popularity with dynamic graph neural network (DGNN) models achieves remarkable performance. However, DGNN models still suffer from tackling sparse datasets where most users are inactive. Therefore, we propose a reformative temporal graph network, named semantics-enhanced temporal graph network (STGN), which attaches extra semantic information into the user-content bipartite graph and could better leverage implicit relationships behind the superficial topology structure. On top of that, we customize its temporal and structural learning modules to further boost the prediction performance. Specifically, in order to efficiently aggregate the diversified semantics that a content might possess, we design a user-specific attention (UsAttn) mechanism for the temporal learning. Unlike the attention mechanism that only analyzes the influence of genres on content, UsAttn also considers the attraction of semantic information to a specific user. Meanwhile, as for the structural learning, we introduce the concept of positional encoding into our attention-based graph learning and novelly adopt a semantic positional encoding (SPE) function, which effectively boost the performance of lightweight algorithms. Finally, extensive simulations verify the superiority of our models and demonstrate their effectiveness in content caching.
引用
收藏
页码:8478 / 8492
页数:15
相关论文
共 50 条
  • [41] Content Popularity Prediction Methods - A Survey
    Nancy, Gnana Amala J.
    Kumar, K.
    PROCEEDINGS OF THE 3RD INTERNATIONAL CONFERENCE ON COMMUNICATION AND ELECTRONICS SYSTEMS (ICCES 2018), 2018, : 749 - 753
  • [42] PPP: Prefix-based Popularity Prediction for Efficient Content Caching in Content-Centric Networks
    Ren, Jianji
    Zhao, Shan
    Sun, Junding
    Li, Ding
    Wang, Song
    Jia, Zongpu
    COMPUTER SYSTEMS SCIENCE AND ENGINEERING, 2018, 33 (04): : 259 - 265
  • [43] Spatio-temporal graph attention networks for traffic prediction
    Ma, Chuang
    Yan, Li
    Xu, Guangxia
    TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH, 2024, 16 (09): : 978 - 988
  • [44] Co-Evolving Popularity Prediction in Temporal Bipartite Networks: A Heuristics Based Model
    Abbas, Khushnood
    Hasan, Mohammad Kamrul
    Abbasi, Alireza
    Dong, Shi
    Ghazal, Taher M.
    Abdullah, Siti Norul Huda Sheikh
    Khan, Asif
    Alboaneen, Dabiah
    Ahmed, Fatima Rayan Awad
    Ahmed, Thowiba E.
    Islam, Shayla
    IEEE ACCESS, 2023, 11 : 37546 - 37559
  • [45] Online Popularity Prediction of Video Segments Towards More Efficient Content Delivery Networks
    Tan, Zhiyi
    Hu, Wen
    Zhang, Ya
    Ding, Hao
    2019 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2019,
  • [46] GPOP: Scalable Group-level Popularity Prediction for Online Content in Social Networks
    Hoang, Minh X.
    Dang, Xuan-Hong
    Wu, Xiang
    Yan, Zhenyu
    Singh, Ambuj K.
    PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB (WWW'17), 2017, : 725 - 733
  • [47] Content Popularity Prediction via Federated Learning in Cache-Enabled Wireless Networks
    YAN Yuna
    LIU Ying
    NI Tao
    LIN Wensheng
    LI Lixin
    ZTE Communications, 2023, 21 (02) : 18 - 24
  • [48] Popularity Prediction Caching Using Hidden Markov Model for Vehicular Content Centric Networks
    Yao, Lin
    Wang, Yuqi
    Xia, QiuFen
    Xu, Rui
    2019 20TH INTERNATIONAL CONFERENCE ON MOBILE DATA MANAGEMENT (MDM 2019), 2019, : 533 - 538
  • [49] Content Popularity Prediction in Fog Radio Access Networks: A Federated Learning Based Approach
    Wu, Yuting
    Jiang, Yanxiang
    Bennis, Mehdi
    Zheng, Fuchun
    Gao, Xiqi
    You, Xiaohu
    ICC 2020 - 2020 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2020,
  • [50] Federated Learning-Based Content Popularity Prediction in Fog Radio Access Networks
    Jiang, Yanxiang
    Wu, Yuting
    Zheng, Fu-Chun
    Bennis, Mehdi
    You, Xiaohu
    IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2022, 21 (06) : 3836 - 3849