Research Article
Hybrid Human-Artificial Intelligence Enabled Edge Caching Based on Interest Evolution
@INPROCEEDINGS{10.1007/978-3-030-89814-4_8, author={Zhidu Li and Fuxiang Li and Dapeng Wu and Honggang Wang and Ruyan Wang}, title={Hybrid Human-Artificial Intelligence Enabled Edge Caching Based on Interest Evolution}, proceedings={Mobile Multimedia Communications. 14th EAI International Conference, Mobimedia 2021, Virtual Event, July 23-25, 2021, Proceedings}, proceedings_a={MOBIMEDIA}, year={2021}, month={11}, keywords={Edge caching User interest evolution Group interest Cache hit rate User hit rate}, doi={10.1007/978-3-030-89814-4_8} }- Zhidu Li
Fuxiang Li
Dapeng Wu
Honggang Wang
Ruyan Wang
Year: 2021
Hybrid Human-Artificial Intelligence Enabled Edge Caching Based on Interest Evolution
MOBIMEDIA
Springer
DOI: 10.1007/978-3-030-89814-4_8
Abstract
How to cache appropriable contents for users from huge amount of candidates is a challenge in edge caching network. To address this challenge, this paper studies an edge caching scheme based on user interest, where an interest extraction and evolution network is developed. Specifically, the input features are first classified and embedding. The user interest is then mined and modeled according to the user historical behaviors with the gated recurrent unit network. Thereafter, the user interest evolution process is studied by analyzing the impact of the previous interests on the current interest through an attention mechanism. The group interest model is further studied by merging user interest evolution and social relationships among contents, based on which edge caching scheme is obtained. The effectiveness of the proposed scheme is finally validated by extensive experiments with a real-world dataset. The analysis in this paper sheds new light on edge content caching from user interest evolution perspective.
