Table 2. Characteristics of the surveyed datasets for fake news detection.

Dataset	News domain	Application purpose	Type of disinformation	Language	Size	News content type	Rating scale	Media platform	Spontaneity	Availability	Extraction time
Yelp dataset (Barbado, Araque & Iglesias, 2019)	Technology	Fake detection	Fake reviews	English	18912 reviews	Text	2 values	Mainstream	Yes	Yes	No
PHEME dataset (Zubiaga et al., 2016)	Society, politics	Rumor detection	Rumors	English and German	330 rumorous conversations and4842 tweets overall	Text	3 values	Social media (Twitter)	Yes	Yes	No
CREDBANK (Mitra & Gilbert, 2015)	Society	Veracity classification	Rumors	English	60 million streaming tweets	Text	5 values	Social media (Twitter)	Yes	Yes	Yes (October 2014 - February 2015)
BuzzFace (Santia & Williams, 2018)	Politics, society	Veracity classification	Fake news articles	English	2263 news	Text	4 values	Social media (Facebook)	Yes	Yes	Yes (September 2016)
FacebookHoax (Tacchini et al., 2017)	Science	Fake detection	Hoaxes	English	15500 posts	Text	2 values	Social media (Facebook)	Yes	Yes	Yes (July 2016 - December 2016)
LIAR (Wang, 2017)	Politics	Fake detection	Fake news articles	English	12836 short statements	Text	6 values	Mainstream + social media (Facebook, Twitter)	Yes	Yes	Yes (2007-2016)
Fact checking dataset (Vlachos & Riedel, 2014)	Politics, society	Fact checking	Fake news articles	English	221 statements	Text	5 values	Mainstream	Yes	Yes	No
FEVER (Thorne et al., 2018)	Society	Fact checking	Fake news articles	English	185,445 claims	Text	3 values	Mainstream	No	Yes	No
EMERGENT (Ferreira & Vlachos, 2016)	Society, technology	Rumor detection	Rumors	English	300 claims, and 2,595 associated article headlines	Text	3 values	Mainstream + social media (Twitter)	Yes	Yes	No
FakeNewsNet (Shu et al., 2018)	Society, politics	Fake detection	Fake news articles	English	422 news	Text, images	2 values	Mainstream + social media (Twitter)	Yes	Yes	No
Benjamin Political News Dataset (Horne & Adali, 2017)	Politics	Fake detection	Fake news articles	English	225 stories	Text	3 values	Mainstream	Yes	Yes	Yes (2014-2015)
Burfoot Satire News Dataset (Burfoot & Baldwin, 2009)	Politics, economy, technology, society	Fake detection	Satire	English	4,233 newssamples	Text	2 values	Mainstream	Yes	Yes	No
BuzzFeed News dataset (Horne & Adali, 2017)	Politics	Fake detection	Fake news articles	English	2,283 news samples	Text	4 values	Social media (Facebook)	Yes	Yes	Yes (2016-2017)
MisInfoText dataset (Torabi & Taboada, 2019)	Society	Fact checking	Fake news articles	English	1,692 news articles	Text	4 values for BuzzFeed and 5 values for Snopes	Mainstream	Yes	Yes	No
Ott et al.’s dataset (Ott et al., 2011)	Tourism	Fake detection	Fake reviews	English	800 reviews	Text	2 values	Social media (TripAdvisor)	No	Yes	No
FNC-1 dataset (Riedel et al., 2017)	Politics, society, technology	Fake detection	Fake news articles	English	49972 articles	Text	4 values	Mainstream	Yes	Yes	No
Spanish fake news corpus (Posadas-Durán et al., 2019)	Science, Sport, Economy, Education, Entertainment, Politics, Health, Security, Society	Fake detection	Fake news articles	Spanish	971 news	Text	2 values	Mainstream	Yes	Yes	Yes (January 2018 - July 2018)
Fake_or_real_news (Dutta et al., 2019)	Politics, society	Fake detection	Fake news articles	English	6,337 articles	Text	2 values	Mainstream	Yes	Yes	No
TSHP-17 (Rashkin et al., 2017)	Politics	Fact checking	Fake news articles	English	33,063 articles	Text	6 values for PolitiFact and 4 values for unreliable sources	Mainstream	Yes	Yes	No
QProp (Barrón-Cedeno et al., 2019)	Politics	Fact checking	Fake news articles	English	51,294 articles	Text	2 values	Mainstream	Yes	Yes	No
NELA-GT-2018 (Nørregaard, Horne & Adalı, 2019)	Politics	Fake detection	Fake news articles	English	713000 articles	Text	2 values	Mainstream	Yes	Yes	Yes(February 2018 - November 2018)
TW_info (Jang, Park & Seo, 2019)	Politics	Fake detection	Fake news articles	English	3472 articles	Text	2 values	social media (Twitter)	Yes	Yes	Yes (January 2015 - April 2019
FCV-2018 (Papadopoulou et al., 2019)	Society	Fake detection	Fake news content	English, Russian, Spanish, Arabic, German, Catalan, Japanese, and Portuguese	380 videos and 77258 tweets	Videos, text	2 values	social media (YouTube, Facebook, Twitter)	Yes	Yes	Yes (April 2017 - July 2017
Verification Corpus (Boididou et al., 2018)	Society	Veracity classification	Hoaxes	English, Spanish, Dutch, French	15629 posts	Text, images, videos	2 values	social media (Twitter)	Yes	Yes	Yes (2012-2015)
CNN / Daily Mail summarization dataset (Jwa et al., 2019)	Politics, society, crime, sport, business, technology, health	Fake detection	Fake news articles	English	287000 articles	Text	4 values	Mainstream	Yes	Yes	Yes (April 2007 - April 2015
Zheng et al.’s dataset (Zheng et al., 2018)	Society	Clickbait detection	Clickbait	Chinese	14922 headlines	Text	2 values	Mainstream + social media (Wechat)	Yes	Yes	No
Tam et al.’s dataset (Tam et al., 2019)	Politics, technology, science, crime, fraud and scam, fauxtography	Rumor detection	Rumors	English	1022 rumors and 4 million tweets	Text	2 values	social media (Twitter)	Yes	Yes	Yes (May 2017 - November 2017)