Multi-dimensional Data-driven Mobile Edge Caching with Dynamic User Preference
Mengge Liu, Dapeng Li, Haitao Zhao, Xiaoming Wang, Rui Jiang · 2020
Mobile proactive edge caching aims to boost the network throughput and improve user experience by leveraging the user-behavior-related information. Generally, the most popular contents need to be identified and cached to fully exploit edge storage capacity. The cache hit rate is often used to evaluate the popularity of the cached content. In this spirit, the content popularity in terms of time and space has been considered from the perspective of the whole regional users. However, the hit action cannot reflect the efficiency of the cache strategy unilaterally. The satisfaction information (i.e., the access time or scores to the content) of users should be used to improve the intelligent cache strategy. Motivated by this observation, this paper proposes a caching scheme that jointly considers the cache evaluation both from the perspectives of the hit action and satisfaction of users. We use real data set to construct the simulations. The experimental results show that the overall cache hit rate is better than that of the existing strategy, and, also, shows a higher user satisfaction score to the cached content.