research question

Youtube, Baidu and Bilibili: to which category do the videos about North Korean defectors belong?

v8svg-us REPORTAGE INTERVIEW AMATEUR VLOG EDU TV SHOW SPEECH DOC NEWS 6 5 3 2 1 44 15 22 INTERVIEW MOVIE EXTRACT NEWS REPORTAGE TRAILER SPEECH REALITY SHOW 7 3 5 3 1 1 1 USA YOUTUBE Number of uploads divided by category (first 100 results) from Nov 2016 to Nov 2017 Hover to highlight CHN Number of uploads divided by category (21 results) from Nov 2016 to Nov 2017 Hover to highlight BAIDU VIDEO AND BILIBILI

Description

The first step to better understand the kind of video contents people are sharing and watching about North Korean defectors is to get familiar with the categories they belong to.

The different categories express the different levels of the discussion, the directions they want to follow and the way they may speak when relating to North Korean defectors. For instance a Reportage will be using a totally different tone from a piece of News and, of course, there will be two different points of view about the topic.

Among all the U.S. videos the hugest amount belongs to the News category, but the situation is completely different for Chinese videos where the most shared category is Movie extract. Again two different ways to relate to the topic and to take part to the discussion.

Protocol

prot8 - iterations: 5- time range: from Nov 2016 to Nov 2017- ranked by: ratings - excluding all the videos before 1 Nov 2016 - considering the first 100 rows- manual labelling of: categories, users To wich category do the videos about North Koreandefectors belong? QUERY RAWGRAPHS TREE MAP Illustrator(refining) RAWGRAPHS TREE MAP Illustrator(refining) VISUALIZATION 250results Excel BBEdit(cleaning) 15results 6results Excel(manual labelling) Excel(manual cleaning+manual labeling) English translation YTDT Video List U.S. NORTH KOREANDEFECTORS 脱北 CHINA Bilibili Youtube Baidu video CORPUS DEFINITION

Youtube.com.us is the source we used to analyze the U.S. videos; on the other hand Baidu videos and Bilibili provided data about Chinese videos. To deepen this specific field the query we used for each source was the same: North Korean defector. We set the time frame from 1 November 2016 to November 2017 in order to collect the most recent insights and to maintain a timing coherence with all the other steps composing the process.

To complete the dataset all the videos were manually labeled into the different categories and this process took into account all the descriptions, video contents and titles.

Data

Timestamp: 11/2016 - 11/2017

Data source: Youtube, Baidu Video,Bilibili

The excel file is organized in information about ranking, publishing date, URL, title, description, topic, view, like, dislike, comments, category and user for each video. There are two folders: one for the U.S. videos analysis and the other for the Chinese part.