Is Data Collection through Twitter Streaming API Useful for Academic Research?

Câmpan, Alina; Atnafu, Tobel; Truta, Traian Marius; Nolan, Joseph R.

doi:10.1109/bigdata.2018.8621898

Cited by 35 publications

(22 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reason for this disproportionate attention may lie in the simplicity of gathering data from Twitter. Twitter enables the downloading of thousands of posts using its official application programming interface [73], whereas Facebook and Instagram closed their application programming interfaces in 2018, thus preventing the automatic downloading of publicly available data from these platforms to protect users' data against inappropriate use [74]. This was in response to the Cambridge Analytica data misuse scandal [75].…”

Section: Principal Findingsmentioning

confidence: 99%

Characteristics of Antivaccine Messages on Social Media: Systematic Review

Wawrzuta¹,

Jaworski²,

Gotlib³

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

Background Supporters of the antivaccination movement can easily spread information that is not scientifically proven on social media. Therefore, learning more about their posts and activities is instrumental in effectively reacting and responding to the false information they publish, which is aimed at discouraging people from taking vaccines. Objective This study aims to gather, assess, and synthesize evidence related to the current state of knowledge about antivaccine social media users’ web-based activities. Methods We systematically reviewed English-language papers from 3 databases (Scopus, Web of Science, and PubMed). A data extraction form was established, which included authors, year of publication, specific objectives, study design, comparison, and outcomes of significance. We performed an aggregative narrative synthesis of the included studies. Results The search strategy retrieved 731 records in total. After screening for duplicates and eligibility, 18 articles were included in the qualitative synthesis. Although most of the authors analyzed text messages, some of them studied images or videos. In addition, although most of the studies examined vaccines in general, 5 focused specifically on human papillomavirus vaccines, 2 on measles vaccines, and 1 on influenza vaccines. The synthesized studies dealt with the popularity of provaccination and antivaccination content, the style and manner in which messages about vaccines were formulated for the users, a range of topics concerning vaccines (harmful action, limited freedom of choice, and conspiracy theories), and the role and activity of bots in the dissemination of these messages in social media. Conclusions Proponents of the antivaccine movement use a limited number of arguments in their messages; therefore, it is possible to prepare publications clarifying doubts and debunking the most common lies. Public health authorities should continuously monitor social media to quickly find new antivaccine arguments and then create information campaigns for both health professionals and other users.

show abstract

Section: Principal Findingsmentioning

confidence: 99%

Characteristics of Antivaccine Messages on Social Media: Systematic Review

Wawrzuta¹,

Jaworski²,

Gotlib³

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…This sample comprises a sample of 22k original tweets and 1.73m retweets, reflecting the prevalence of retweets about this topic. Data were collected via an application programming interface (API) (Campan et al, 2019). We used Twitter's Premium API to access historical data with a monthly cap of 1.25 million tweets.…”

Section: Collecting and Engineering Twitter Datamentioning

confidence: 99%

Using Twitter Data to Monitor Immigration Sentiment

Rowe¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The detection of the duplicate fake account on twitter using features based analysis, SVM is also used. SVM basically divide using the vector, helps in developing the decisions using hyper planes [19]. It explains the close data points from the hyperplane, the information retrieval like data points against the features selected for duplicate fake accounts.…”

Section: Support Vector Machinementioning

confidence: 99%

Feature Based Analysis on the Duplicate Fake Account on Twitter Using Intelligent Mechanism

Alhariri¹

2021

ITII

View full text Add to dashboard Cite

The duplicate fake accounts are detected in this work the data from the social media platform is accessed. The platform choose to use the analysis on social media platform is selected as twitter. The twitter data is accessed using Twitter API, with using some selected features that remain the most appropriate regarding the reason of duplicate fake account. The feature based analysis is compared using machine learning techniques, Random Forest, Decision Tree, and SVM. The performance is further analyzed based on accuracy SVM performed 93.3% accuracy, where decision tree performed as 89.0% and random forest performed as 85.5%. The better performance observed using feature-based analysis is of SVM.

show abstract

Is Data Collection through Twitter Streaming API Useful for Academic Research?

Cited by 35 publications

References 2 publications

Characteristics of Antivaccine Messages on Social Media: Systematic Review

Characteristics of Antivaccine Messages on Social Media: Systematic Review

Using Twitter Data to Monitor Immigration Sentiment

Feature Based Analysis on the Duplicate Fake Account on Twitter Using Intelligent Mechanism

Contact Info

Product

Resources

About