With the rapid proliferation of the Internet and mobile devices, vast amounts of user-generated content has been accumulated through social network services, and massive amounts of news continues to be created and posted online. In this paper, we extract the characteristics of online news and data from social network services, including the differences and similarities between them. We found the following differences: First, the news responds to official events but content on social network services is related to personal interests. Second, the news is continually related to a specific issue or set of issues, whereas topics of conversation change daily in social network services. Third, items from the news can be identified with a single keyword in searches, whereas more keywords are needed to extract the desired information from social network services. At the same time, we found that the words mentioned in both the news and on social network services were similar, and both were used for commercial purposes. Our analysis revealed that the news is related to the keyword generally, uses same words repeatedly, and its range of topics is narrow and public. On the contrary, social network services are not related to the keyword often, and their range of topics is wide and personal. Furthermore our analysis showed that the ranking algorithm improves the topic detection rate and catches the topic quickly. This paper provides useful information to better understand the characteristics of online news and data from social network services.
Bibliographical noteFunding Information:
This work was supported by the National Research Foundation of Korea though the Korean government under Grant 2016R1E1A2A01954003.
© 2013 IEEE.
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Materials Science(all)