國立虎尾科技大學 |

Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks./
Author:	Mehta, Swapneel.
Description:	1 online resource (122 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-04, Section: B.
Contained By:	Dissertations Abstracts International85-04B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9798380620567

Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks.
Mehta, Swapneel.

Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks. - 1 online resource (122 pages)

Source: Dissertations Abstracts International, Volume: 85-04, Section: B.

Thesis (Ph.D.)--New York University, 2023.

Includes bibliographical references

We assess the efficacy of interventions that can limit the abuse of social networks to spread misleading information online. First, we identify the causal effects of a popular content moderation practice on mainstream social platforms including Twitter, Reddit, Instagram and Facebook (now, Meta), aimed at curbing the spread of misleading information. Our work shows that Twitter's application of warning labels to Donald Trump's misleading tweets exhibited a statistically significant Streisand effect, thereby causing an increase in engagement with these tweets on Twitter, and had varying effects on posts linking to these tweets shared on Facebook, Reddit, and Instagram. Second, we provide causal estimates to show that a class of visibility related interventions arising from platform downtime can cause a reduction in the frequency of posting low-quality news on Reddit while simultaneously increasing the spread of high-quality news. Third, we assess the building blocks of social networks, presenting a method to audit how different ranking and recommendation algorithms are susceptible to the amplification of online harms. We design a virtual testbed to 'red team' algorithmic vulnerabilities on social networks, using forward generative models of user behavior calibrated using a dataset of millions of posts from Reddit. Collectively, we contribute to the literature on assessing the cross-platform causal effects of content moderation practices, present a novel research design identifying a means reduce the sharing of low-quality news, and provide a simulation-based framework to improve content distribution algorithms on social networks.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380620567Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Causal inferenceIndex Terms--Genre/Form:

554714
Electronic books.

Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks.
LDR:03049ntm a22003977 4500 001 1147444
005 20240918072811.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380620567
035 $a (MiAaPQ)AAI30631775
035 $a AAI30631775
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Mehta, Swapneel. $3 1473181
245 1 0 $a Towards Informed Interventions to Limit the Effects of Misleading Information on Social Networks.
264 0 $c 2023
300 $a 1 online resource (122 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-04, Section: B.
500 $a Advisor: Bonneau, Richard;Ranganath, Rajesh.
502 $a Thesis (Ph.D.)--New York University, 2023.
504 $a Includes bibliographical references
520 $a We assess the efficacy of interventions that can limit the abuse of social networks to spread misleading information online. First, we identify the causal effects of a popular content moderation practice on mainstream social platforms including Twitter, Reddit, Instagram and Facebook (now, Meta), aimed at curbing the spread of misleading information. Our work shows that Twitter's application of warning labels to Donald Trump's misleading tweets exhibited a statistically significant Streisand effect, thereby causing an increase in engagement with these tweets on Twitter, and had varying effects on posts linking to these tweets shared on Facebook, Reddit, and Instagram. Second, we provide causal estimates to show that a class of visibility related interventions arising from platform downtime can cause a reduction in the frequency of posting low-quality news on Reddit while simultaneously increasing the spread of high-quality news. Third, we assess the building blocks of social networks, presenting a method to audit how different ranking and recommendation algorithms are susceptible to the amplification of online harms. We design a virtual testbed to 'red team' algorithmic vulnerabilities on social networks, using forward generative models of user behavior calibrated using a dataset of millions of posts from Reddit. Collectively, we contribute to the literature on assessing the cross-platform causal effects of content moderation practices, present a novel research design identifying a means reduce the sharing of low-quality news, and provide a simulation-based framework to improve content distribution algorithms on social networks.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Political science. $3 558774
650 4 $a Web studies. $3 1148502
653 $a Causal inference
653 $a Deep learning
653 $a Misinformation
653 $a Policy
653 $a Social media
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0615
690 $a 0646
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a New York University. $b Center for Data Science. $3 1468521
773 0 $t Dissertations Abstracts International $g 85-04B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30631775 $z click for full text (PQDT)