Multimedia cloud content distribution based on interest discovery and integrated utility of user

摘要

With the rapid development of intelligent devices and the improvement of multimedia services, intelligent devices have become the mainstream source of getting multimedia service. However, there are still some problems influencing user experience of multimedia content acquisition, like limited bandwidth, large user visits and server distribution imbalance. To address the abovementioned issues, this paper proposes a content distribution method based on interest discovery and integrated utility of user in multimedia cloud. In proposed method, we use the improved feature extraction algorithm to extract the user's interest features, calculate the similarity of users, and then the users with adjacent region and similar service interest are categorized into a group, in which the service users and non-service users are separated. Aiming at the selection problem of service users in the user group, the paper has jointly considered user physical performance, user selfish behaviors and user reputation, which all are used to constitute the integrated utility value. In order to minimize content distribution time and user cost, the improved extremum disturbed particle swarm optimization algorithm is used for determining service user number. The results show that the proposed method can significantly improve overall system performance and reduce the total cost of multimedia cloud users.