When Machine Learning Meets Media Cloud:Architecture,Application and Outlook  

作　　者：JIN Yichao WEN Yonggang 

机构地区：[1]School of Conlputer Science and Engineering,Nanyang Technological University,639798,Singapore

出　　处：《ZTE Communications》2018年第3期30-39,共10页中兴通讯技术（英文版）

摘　　要：Nowadays, media cloud and machine learning have become two hot research domains. On the one hand, the increasing user de- mand on multimedia services has triggered the emergence of media cloud, which uses cloud computing to better host media servic- es. On the other hand, machine learning techniques have been successfully applied in a variety of multimedia applications as well as a list of infrastructure and platform services. In this article, we present a tutorial survey on the way of using machine learning techniques to address the emerging challenges in the infrastructure and platform layer of media cloud. Specifically, we begin with a review on the basic concepts of various machine learning techniques. Then, we examine the system architecture of media cloud, focusing on the functionalities in the infrastructure and platform layer. For each of these function and its corresponding challenge, we further illustrate the adoptable machine learning based approaches. Finally, we present an outlook on the open issues in this intersectional domain. The objective of this article is to provide a quick reference to inspire the researchers from either machine learning or media cloud area.Nowadays, media cloud and machine learning have become two hot research domains. On the one hand, the increasing user de- mand on multimedia services has triggered the emergence of media cloud, which uses cloud computing to better host media servic- es. On the other hand, machine learning techniques have been successfully applied in a variety of multimedia applications as well as a list of infrastructure and platform services. In this article, we present a tutorial survey on the way of using machine learning techniques to address the emerging challenges in the infrastructure and platform layer of media cloud. Specifically, we begin with a review on the basic concepts of various machine learning techniques. Then, we examine the system architecture of media cloud, focusing on the functionalities in the infrastructure and platform layer. For each of these function and its corresponding challenge, we further illustrate the adoptable machine learning based approaches. Finally, we present an outlook on the open issues in this intersectional domain. The objective of this article is to provide a quick reference to inspire the researchers from either machine learning or media cloud area.

关 键 词：machine learning media cloud 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]