A Cross-Platform Collection of Social Network Profiles

Veiga, Maria Han; Eickhoff, Carsten

doi:10.1145/2911451.2914666

Cited by 9 publications

(4 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Finally, we compute the average tweeting rate for every user as the ratio of total number of lifetime tweets over the number of days since the account was created. To exclude what are likely to be automated accounts in both rumor and control datasets, we retain users with an average tweeting rate less than or equal to 24 tweets per day (following posting activity thresholds such as in [43,68]). Applying this criteria, we discard 6,463 (26%) users from rumor and 144,904 (31%) users from control sets.…”

Section: User Selectionmentioning

confidence: 99%

Fake Cures

Ghenai

Mejova

2018

Proc. ACM Hum.-Comput. Interact.

View full text Add to dashboard Cite

Social media's unfettered access has made it an important venue for health discussion and a resource for patients and their loved ones. However, the quality of the information available, as well as the motivations of its posters, has been questioned. This work examines the individuals on social media that are posting questionable health-related information, and in particular promoting cancer treatments which have been shown to be ineffective (making it a kind of misinformation, willful or not). Using a multi-stage user selection process, we study 4,212 Twitter users who have posted about one of 139 such "treatments", and compare them to a baseline of users generally interested in cancer. Considering features capturing user attributes, writing style, and sentiment, we build a classifier which is able to identify users prone to propagate such misinformation at an accuracy of over 90%, providing a potential tool for public health officials to identify such individuals for preventive intervention. A. Ghenai & Y. Mejovaimposed on often profit-seeking websites, social media provides a dynamic forum for propagating possible medical misinformation [31]. Recent rise in vaccine hesitancy has been linked to an active movement on Twitter, promoting conspiratorial thinking and mistrust in the government [65]. Image sharing platforms such as Flickr and Instagram have become battlegrounds between the pro-anorexia movement and physicians attempting to intervene [13,92]. Uncertainty surrounding infectious disease outbreaks, such as the Zika epidemic of 2016, yielded rumors and speculations about its causes, preventive measures, and consequences [23,34].In this study we turn to the individuals sharing questionable medical information on Twitter, in particular cancer treatments which have been medically proven to be ineffective. Having around 336 million monthly active users in the first quarter of 2018 1 , Twitter is one of the largest social media websites expressly dedicated to the sharing of information, including that on cancer. Compiling hundreds of thousands of tweets on 139 queries spanning acupuncture, cinnamon, reflexology, and vitamin C, we apply strict selective criteria employing human/organization classification [61], name dictionaries, usage thresholds, and crowdsourced relevance refinement resulting in 4,212 users, which we then compare to those mentioning cancer in general from a previous study [70]. Employing previous research on rumor detection, we characterize these users in multi-faceted feature spaces, encompassing user attributes, linguistic style, sentiment, and post timing. We find users who have a more sophisticated language, who are interested in cancer, but who are not personally involved with the illness. We build a logistic regression model which, out of Twitter users mentioning cancer, is able to identify those who will eventually post a piece of misinformation with a high level of accuracy.Misinformation on social media is an urgent issue, and even more so in the health field. This paper is one of th...

show abstract

Section: User Selectionmentioning

confidence: 99%

Fake Cures

Ghenai

Mejova

2018

Proc. ACM Hum.-Comput. Interact.

View full text Add to dashboard Cite

show abstract

“…There is growing interest in cross-platform research. Methods to facilitate cross-platform analysis have focused on similarities in usernames and user metadata, which may capture only a subset of activity across platforms (Han Veiga and Eickhoff 2016; Malhotra et al 2012; Goga et al 2013; Zafarani and Liu 2009). This content-focused method offers an alternative approach for cross-platform research in social science, one that can capture the spread of ideas and information relevant to many areas such as the spread of misinformation and social mobilization.…”

Section: Discussionmentioning

confidence: 99%

How Information Flows from the World to China

Schaefer²,

Park

et al. 2022

The International Journal of Press/Politics

View full text Add to dashboard Cite

Government censorship—internet shutdowns, blockages, firewalls—impose significant barriers to the transnational flow of information despite the connective power of digital technologies. In this paper, we examine whether and how information flows across borders despite government censorship. We develop a semi-automated system that combines deep learning and human annotation to find co-occurring content across different social media platforms and languages. We use this system to detect co-occurring content between Twitter and Sina Weibo as Covid-19 spread globally, and we conduct in-depth investigations of co-occurring content to identify those that constitute an inflow of information from the global information ecosystem into China. We find that approximately one-fourth of content with relevance for China that gains widespread public attention on Twitter makes its way to Weibo. Unsurprisingly, Chinese state-controlled media and commercialized domestic media play a dominant role in facilitating these inflows of information. However, we find that Weibo users without traditional media or government affiliations are also an important mechanism for transmitting information into China. These results imply that while censorship combined with media control provide substantial leeway for the government to set the agenda, social media provides opportunities for non-institutional actors to influence the information environment. Methodologically, the system we develop offers a new approach for the quantitative analysis of cross-platform and cross-lingual communication.

show abstract

“…They extracted representative features from these sources to characterize users with various views. Han Veiga and Eickhoff (2016) collected data from three different social media platforms (Twitter, Foursquare and Instagram) and have tried to create online footprints of 850 users descriptively. Similarly, Farseev et al (2015) performed a comprehensive user profile learning from different views by using data from Twitter, Foursquare, Instagram and Facebook.…”

Section: Literature Reviewmentioning

confidence: 99%

Do Twitter phenomena check-in popular venues on Foursquare too?

Bozanta

Kutlu

2018

IDD

View full text Add to dashboard Cite

Purpose The purpose of this study is to figure out the visiting behaviors of the users who have different characteristics on Twitter. Design/methodology/approach The visit history of users who share their Foursquare check-ins on Twitter and the characteristics of visited venues (category, check-in count, tip count, like count, rating, and price tier) was collected with Foursquare API. In addition, the number of followers, friends, tweets and favorite-count were collected via Twitter API. First, users were clustered according to their Twitter related attributes. After that, profiling was applied on clusters according to the characteristics of the venues that were visited by the users. Findings Clustering analysis generated three clusters, namely, ordinary, talkative and popular. For each cluster, the visited venues were investigated according to the price classification, check-in, like, tip counts and the categories. The users in ordinary class prefer cheaper venues rather than talkative and popular users. On the other hand, popular users prefer the venues with the highest average number of check-ins, likes and tip counts. The top two categories for all clusters are cafe and shopping mall. Originality/value This study differentiates from the other studies in the literature by examining the data from Twitter with clustering and profiling these clusters with Foursquare data to understand venue preferences of Twitter users having various characteristics. The findings of this study will provide new insights for business owners to understand the customers more comprehensively and design better marketing strategies.

show abstract

A Cross-Platform Collection of Social Network Profiles

Cited by 9 publications

References 6 publications

Fake Cures

Fake Cures

How Information Flows from the World to China

Do Twitter phenomena check-in popular venues on Foursquare too?

Contact Info

Product

Resources

About