Bloomberg TV talkes to Ontotext’s CEO Atanas Kiryakov about fake news, human behavior analysis and prediction, filtering and serving content with semantic technology and more.
Our study suggests that the #BRexit referendum is anything but getting to a close, with a clear preference among Twitter users for leaving the EU. Ontotext analyzed over 1.5 million tweets related to the referendum up until 13 May 2016.
The tweets were identified by common referendum-related terms, such as #brexit; #StrongerIn; “uk eu vote”, and then analyzed for the use of hashtags, cited, and general sentiment for leaving or staying in the EU.
94.4% of the selected tweets were in English (i.e., English is the language predicted by Twitter’s algorithm for language detection). Below is the distribution of the other languages:
You can see more details about the Twitter data as well as links to actual data files are available in the full version of the report.
We looked at users with the biggest number of posts in our dataset, also considering their polarity. At a first glance, it appears that most of them are not public figures or media companies (see the figure below).
Propagandistic Twitter users are the most influential, in terms of the number of posts and the number of retweets. They are dominating the top of the list. It would be interesting to find out which are the actors behind these propagandistic users – whether they are politicians, parties, organizations, etc.
We went further and looked at the users mentioned in the tweets by frequency and by the polarity of the context. The number of retweets a Twitter user receives is a pretty strong gauge of their influence on the wider conversation. According to this analysis, the top five most re-tweeted Twitter users on the topic of the EU referendum are all Leave campaigners.
The same analysis as above was conducted for publishing domains frequently cited in the tweets. E.g., a tweet can cite an article from bbc.co.uk in support of their statement.
The big question is whether someone’s activity on Twitter forms an accurate indication of their likelihood to go out and vote on the day. We’ll find out soon.