RT @MishaTeplitskiy: Twitter's Sample API allegedly gave random sample of Tweets. But it wasn't fully random, it took Tweets at particular…
RT @MishaTeplitskiy: Twitter's Sample API allegedly gave random sample of Tweets. But it wasn't fully random, it took Tweets at particular…
Twitter's Sample API allegedly gave random sample of Tweets. But it wasn't fully random, it took Tweets at particular known time intervals. So one could Tweet at particular times to get into the sample, creating weird marketing/injection opportunities http
@mattsclancy @straightedge verrrry long-con injection attack, duh (not infeasible, it turns out: https://t.co/udw0clBBa5 )
How does Twitter API sampling work? Randomly, according to Twitter. Turns out every tweet has a TweetID (a unique 64-bit representation containing timestamp in milliseconds, data center, worker thread, sequence n.). This allowed to manipulate the sample. h
Dec 2018 article about Twitter sample feed manipulation. "To make Twitter a reliable data source for future research, especially in the social sciences, we must find ways to create scientifically sound samples from these data streams." Tall order. https://
@mcpowr I wonder what this looks like for scheduled tweets: https://t.co/JwmHlvMaHn
On a related note, here is a great article on the complications of using Twitter's sample API for research: https://t.co/VEbClf4ml0
and if you're smart/motivated, you can use the non-randomness to tamper with the sample https://t.co/udw0cljZLv
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
RT @JurgenPfeffer: NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Ph…
NOT SO BAD AFTER ALL! :o) We figured out how to do the extended call to get about the same data as with the old API. Phew, real-time science is nerve-racking. Again, thanks to @fredmorstatter @mehrnoom More info: https://t.co/u3gtIigTjg https://t.co/fx
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
Tampering with Twitter’s Sample API Tampering with Twitter’s Sample API EPJ #datascience volume 7, Article number: 50 (2018) Cite this article Article met .. https://t.co/anEoi4cRfn https://t.co/4W2HyNCam9
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: @dgarcia_eu @fredmorstatter @mehrnoom Correct! Switching to the 1ms sampling windows is the easiest possible solution to…
RT @jasonbaumgartne: .@luca have you seen this tweet: https://t.co/UaeC9TlJ21 I am trying to confirm @JurgenPfeffer's findings but I am no…
.@luca have you seen this tweet: https://t.co/UaeC9TlJ21 I am trying to confirm @JurgenPfeffer's findings but I am not seeing this change on my sample stream.
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
@dgarcia_eu @fredmorstatter @mehrnoom Correct! Switching to the 1ms sampling windows is the easiest possible solution to avoid major manipulation potential we showed before (https://t.co/ryEIa9pkVG). 1ms windows are harder to manipulate, but not impossible
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
RT @JurgenPfeffer: Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms win…
@pushkalagarwa @NirmalSivaraman
Twitter changed 1% Sample data! 1st info: Dramatically less meta-data (NO sender info, just id). Sampling now w/ 1ms window (old: 10ms). Not good for Science! #CSS #Twitter Thanks: @fredmorstatter @mehrnoom Background: Our 2018 paper on Twitter sampling:
RT @JurgenPfeffer: A big shout-out to @rhinelaendr (Dennis Kergl) et al. for their work on revealing much of Twitter's Tweet handling logis…
Tampering with Twitter’s Sample API https://t.co/JuZFPIht5l
Toller Talk zur Einführung in die Netzwerkanalyse @luca ! Frage und Warnung: Wie gehst du mit potentiell möglicher Manipulation vor allem der kostenlosen 1% #Twitter Streaming-Api um? Paper von @JuergenPfeffer hier: https://t.co/8A71H3t8s8 #rp19 #bots
Twitter の Sample API は適切な無作為抽出になっているかどうかの検証。 / Tampering with Twitter's Sample API (2018) https://t.co/4LFhdy1bVm
@jasonbaumgartne This paper by @JurgenPfeffer et al might also be of interest https://t.co/Lyv0tSNMHu
RT @JurgenPfeffer: New article @epj_ds +++ It is easy to tamper with Twitter’s 1% & 10% Sample API +++ Volume of any topic can be manipulat…
RT @JurgenPfeffer: New article @epj_ds +++ It is easy to tamper with Twitter’s 1% & 10% Sample API +++ Volume of any topic can be manipulat…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
#Twitter #TwitterDaten in der #Forschung #FDM
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
RT @GESIS_Panel: #RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of a…
#RecommendedPaper Researchers often use Twitter data in their analysis. This platform says it provides random samples of all tweets, but is the sample actually random? The authors of this paper investigate on this question. https://t.co/ZO0z2FNAUH @Jurgen
RT @alexvespi: New in @epj_ds Tampering with Twitter’s Sample API “due to the nature of Twitter’s sampling mechanism, it is possible to d…
EPJ Data Sci.: Tampering with Twitter’s Sample API https://t.co/jKFn5okVS8
.@amelia_acker
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @ProfessMoravec: feeling much happier about all that hand scraping of every dang #digitalhumanities tweet I did for Disrupting the Digit…
RT @ProfessMoravec: feeling much happier about all that hand scraping of every dang #digitalhumanities tweet I did for Disrupting the Digit…
feeling much happier about all that hand scraping of every dang #digitalhumanities tweet I did for Disrupting the Digital Humanities essay
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @ryanjgallag: Just got around to reading @JurgenPfeffer, @katja_mat and @fredmorstatter's paper on how Twitter's sampling APIs (includi…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
RT @elisa_omodei: Tampering with Twitter’s Sample API | @JurgenPfeffer et al show "how a flaw in the sampling mechanism underpinning Twitte…
Yikes, time for Twitter to develop a less naive sampling scheme