9

I want to know about recently available datasets for fake news analysis

  • 3
    There are methods/algorithms to automatically identify fake news, see e.g. 10.1002/pra2.2015.145052010082. I don't know whether there are ooen datasets. – FuzzyLeapfrog Feb 5 '17 at 20:46
  • So, you want samples of fake news so that you can analyze these news articles, right? – Nicolas Raoul Mar 23 '17 at 7:34
  • I need an annotated dataset with fake and real news articles with their links – Paramie.Jayasinghe Mar 31 '17 at 6:36
5

Buzzfeed News has been doing work on this, and has published data related to fake news, news patterns, and social media patterns on their Github: https://github.com/BuzzFeedNews/everything. Might be a good repo to browse.

| improve this answer | |
2

Here are some of the datasets available for fake news detection:

LIAR dataset: https://www.cs.ucsb.edu/william/data/liar_dataset.zip

BS Detector: https://github.com/bs-detector/bs-detector

| improve this answer | |
1

You should check out the Observatory on Social Media (OSoMe) at Indiana University. The team have been been archiving 10% of public activity on Twitter for the last 10 years. The data isn't directly available to people not affiliated with the University they have a number of algorithms and visualization tools that you can run against the data.

  • They have a service called 'BotSlayer' which you can set up yourself on a free AWS instance and track certain hashtags and key phrases.
  • There is also 'Botometer'which will assess any twitter user name and socre it based on how 'bot-like' it is.
  • Finally, they have a tool called 'Hoaxy' which allows you to visualize the spread of a news or fake-news story across twitter to see which accounts are sharing/re-tweeting it.
| improve this answer | |
0

Kaggle hosts a dataset where the CSV has URL, title, text, and a flag "reliable" or "unreliable"

https://www.kaggle.com/c/fake-news/data

id: unique id for a news article

title: the title of a news article

author: author of the news article text: the text of the article; could be incomplete

label: a label that marks the article as potentially unreliable

1: unreliable

0: reliable

accessing the data requires registration

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.