期刊名称：International Journal of Networking and Computing
印刷版ISSN：2185-2847
出版年度：2017
卷号：7
期号：1
页码：86-104
语种：English
出版社：International Journal of Networking and Computing
摘要：This study considers the problem of selecting a small number of important persons from social media. Skyline query has been utilized for selecting key persons. Based on certain criteria from social media, this query selects persons who are not dominated by any other. Owing to the complex structure of social media, selecting a key person is more complicated and its application is quite different from conventional skyline queries. We need to consider various metrics in the social media. In addition, social media contains massive data, and the data increase is huge. It is collection of online communication channels dedicated to community-based inputs, interactions, content sharing, and collaboration. We use MapReduce framework to speed up the computation and introduce parallelism in the processing. An extensive set of experiments shows that the analysis of social activities, social relationships, and socially shared contents helps finding a key person. The experimental results also confirm the efficiency and scalability of our algorithm on a synthetic dataset.
其他摘要：This study considers the problem of selecting a small number of important persons from social media. Skyline query has been utilized for selecting key persons. Based on certain criteria from social media, this query selects persons who are not dominated by any other. Owing to the complex structure of social media, selecting a key person is more complicated and its application is quite different from conventional skyline queries. We need to consider various metrics in the social media. In addition, social media contains massive data, and the data increase is huge. It is collection of online communication channels dedicated to community-based inputs, interactions, content sharing, and collaboration. We use MapReduce framework to speed up the computation and introduce parallelism in the processing. An extensive set of experiments shows that the analysis of social activities, social relationships, and socially shared contents helps finding a key person. The experimental results also confirm the efficiency and scalability of our algorithm on a synthetic dataset.
