Media Cloud: Massive Open Source Collection of Global News on the Open Web

Roberts, Hal; Bhargava, Rahul; Valiukas, Linas; Jen, Dennis; Malik, Momin M.; Bishop, Cindy Sherman; Ndulue, Emily Boardman; Dave, Aashka; Clark, Justin; Etling, Bruce; Faris, Rob; Shah, Anushka; Rubinovitz, Jasmin; Hope, Alexis; D’Ignazio, Catherine; Bermejo, Fernando; Benkler, Yochai; Zuckerman, Ethan

doi:10.1609/icwsm.v15i1.18127

Cited by 27 publications

(15 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 8: Dynamics of mean centrality measures in the opinion co-occurrence network for conspiracy (red lines) and non-conspiracy opinions (gray lines). The green line shows the news coverage ratios from Media Cloud (Roberts et al 2021). The highlighted area shows a spike in the news coverage, which coincides with a decrease in the centrality of conspiracy opinions.…”

Section: Centrality Dynamics In Opinion Networkmentioning

confidence: 99%

See 1 more Smart Citation

Slipping to the Extreme: A Mixed Method to Explain How Extreme Opinions Infiltrate Online Discussions

Kong

Booth

Bailo

et al. 2022

ICWSM

View full text Add to dashboard Cite

Qualitative research provides methodological guidelines for observing and studying communities and cultures on online social media platforms. However, such methods demand considerable manual effort from researchers and can be overly focused and narrowed to certain online groups. This work proposes a complete solution to accelerate the qualitative analysis of problematic online speech, focusing on opinions emerging from online communities by leveraging machine learning algorithms. First, we employ qualitative methods of deep observation for understanding problematic online speech. This initial qualitative study constructs an ontology of problematic speech, which contains social media postings annotated with their underlying opinions. The qualitative study dynamically constructs the set of opinions, simultaneous with labeling the postings. Next, we use keywords to collect a large dataset from three online social media platforms (Facebook, Twitter, and Youtube). Finally, we introduce an iterative data exploration procedure to augment the dataset. It alternates between a data sampler --- which balances exploration and exploitation of unlabeled data --- the automatic labeling of the sampled data, the manual inspection by the qualitative mapping team, and, finally, the retraining of the automatic opinion classifiers. We present both qualitative and quantitative results. First, we show that our human-in-the-loop method successfully augments the initial qualitatively labeled and narrowly focused dataset and constructs a more encompassing dataset. Next, we present detailed case studies of the dynamics of problematic speech in a far-right Facebook group, exemplifying its mutation from conservative to extreme. Finally, we examine the dynamics of opinion emergence and co-occurrence, and we hint at some pathways through which extreme opinions creep into the mainstream online discourse.

show abstract

Section: Centrality Dynamics In Opinion Networkmentioning

confidence: 99%

“…We also depict the attention dedicated by the Australian news media to the bushfires during the same period. We estimate the latter using the news coverage ratio -the percentage of articles dedicated to the topic over all captured articles in a day -crawled using the Media Cloud (Roberts et al 2021).…”

Section: Centrality Dynamics In Opinion Networkmentioning

confidence: 99%

Slipping to the Extreme: A Mixed Method to Explain How Extreme Opinions Infiltrate Online Discussions

Kong

Booth

Bailo

et al. 2022

ICWSM

View full text Add to dashboard Cite

show abstract

“…We used the an open-source platform for media analysis MediaCloud available at https://mediacloud.org/ (Roberts et al 2021), searching the geographical collection United Kingdom -National, that includes most UK media sources. We performed a Boolean search including all names of the GBD or JSM list and the terms "COVID-19 or coronavirus or pandemic" within the date range 1/1/2020-20/10/2021.…”

Section: Identification Of Uk News Articles Mentioning Gbd or Jsm Nam...mentioning

confidence: 99%

Political polarization in the frequency British newspapers mention scientists with different views on COVID-19

Ghezzi¹,

Giglietto²,

Martinisi³

2023

Preprint

View full text Add to dashboard Cite

Background. Measures against the COVID-19 pandemics have been subject of political polarization also in the United Kingdom. While this has been studied in the broader context of misinformation, the debate around non-pharmacological interventions (NPI) , the and the political implications of different views of their risks and benefits, has not been taken into account with sufficient depth Research question. Our research question was: Are politically-oriented UK newspapers more likely to promote a more, or less, precautionary views toward the use of NPIs?Methods. We used the Mediacloud database to search the names of scientists with different precautionary views (i.e.: signatories of the Great Barrington Declaration and of the John Snow Memorandum) and analyze the frequency with which they are mentioned in newspapers with different political orientation.Results. Scientists with more precautionary positions are more likely to be mentioned in left-leaning newspapers, whereas those with less precautionary views on right-leaning ones. The two groups segregated also in terms of co-authorship of academic papers.Conclusions. Scientists with different opinions in relation to NPIs are mentioned by UK newspapers in a way that is dependent on the political orientation of the newspaper.

show abstract

“…We used the open-source platform for media analysis MediaCloud available at https://mediacloud.org/ [35], searching the geographical collection "U.S. Top Newspapers 2018" or the collection "Italy -National" for the date range 1/1/2020-20/10/2021. In English, we performed seven independent searches of these terms referring to COVID-19 interventions: convalescent plasma, hydroxychloroquine, ivermectin, lockdown, mask, vaccine and vitamin D; these were combined in a Boolean AND search with "covid AND (published OR publication OR journal)".…”

Section: Data Collectionmentioning

confidence: 99%

Reporting on scientific studies in newspaper articles on COVID-19 in Italy and the US: missing citations and information bubbles

Palatta¹,

Macedo²,

Marcacci³

et al. 2023

Preprint

View full text Add to dashboard Cite

Newspapers are a major source of health information, including on COVID-19. Many news articles are based on press releases describing research published in scientific journals. We studied a sample of articles in US and Italian newspapers for their mention of non-pharmacological interventions (face masks and lockdown) and pharmacological ones, both effective and ineffective (convalescent plasma, hydroxychloroquine, ivermectin, vaccines and vitamin D), and whether these were mentioned in a favorable or unfavorable way. We checked for the presence or absence of explicit mentions or links of the primary source of scientific information. Finally, we analyzed whether there was a trend to form “information bubbles” where some opinions around different treatments cluster together.Of 480,819 news in the USA and 767,172 in Italy, vaccines, face masks and lockdown were the most mentioned interventions. Of the pharmacological interventions other than vaccines, ivermectin and hydroxychloroquine were more frequently mentioned in the USA than in Italy (5- and 6-fold, respectively) although, when analyzing a sample of 210 news returned from a search on COVID-19 mentioning a research publication, articles from the USA were less favorable than those in Italy. We also found that the frequency of articles with a negative stance on vaccines was very small, indicating that the main newspapers do not contribute significantly to vaccine hesitancy. There was also evidence of information bubbles, where articles with a favorable or unfavorable view of a non-approved drug had the same stance on other non-approved treatments.Of the 210 news articles analyzed, half specifically mentioned a scientific publication. However, a link (or a complete citation) to the original source was provided in only 16% of Italian newspapers as opposed to 58% of those in the USA. The results highlight the fact that often news stories do not cite the scientific article they are reporting on, thus allowing the reader to verify the original source. This weakness in the citation behavior is particularly evident in Italian newspapers. This study suggests that linking to the original research article, rather than basing the news story on a press release, would improve the trustworthiness of the news and the critical thinking in the reader.

show abstract

Media Cloud: Massive Open Source Collection of Global News on the Open Web

Cited by 27 publications

References 36 publications

Slipping to the Extreme: A Mixed Method to Explain How Extreme Opinions Infiltrate Online Discussions

Slipping to the Extreme: A Mixed Method to Explain How Extreme Opinions Infiltrate Online Discussions

Political polarization in the frequency British newspapers mention scientists with different views on COVID-19

Reporting on scientific studies in newspaper articles on COVID-19 in Italy and the US: missing citations and information bubbles

Contact Info

Product

Resources

About