Hi, everyone. I'm Yang Guanze. I have
learned a lot about social media in the course of IEMS5723, and very happy to
share some thoughts about content analysis with you here.
Social media is becoming increasingly
important in people daily life as the development of the internet and mobile
technology. It’s normal to see people share their feelings and interesting
experience in micro-blog or their circle of friends. And content analysis is a crucial way for the social media to know
how people think and perform in certain condition, and by using that can help
social media improve their users experience.
But the contents which generated by users
are really are too many. According to the investigation, 0.3 billion users of
sina weibo generate 100TB’s text content, which not including comments and
pictures. And the variety of the data of sina weibo is very complex too(Figure 1.). So how to
collect and analyze them?
![]() |
| Figure 1. Variety of Social Network data |
●natural language
●profile/tags
●comments and repost
●relationships of users
Big data are high-volume, high-velocity,
and /or high-variety information assets that require new forms of processing to
enable enhanced decision making, insight discovery and process optimization. As mentioned above, the data in social media are kind of big data which users disclosure them voluntarily.
I think the essential way to process the
data is to analyze the text content. The processing of large data can be seen
as a natural extension of data mining, called the text mining. [1]And after that we
can do the natural language processing(NLP).
Besides the concept which mentioned by our
teacher on class, I think natural language can be divided into two parts:
objective language and subjective language. The former is a statement of fact, and
the latter expressed emotion. We can get people’s thoughts by the extraction of
subjective language, which will help to improve social media’s performance.
Reference:
[1] Wei Li, Tian Tang, Lei Li. Mining. Mining Public Opinions from Chinese Social Media.
