On Microtargeting Socially Divisive Ads: Mahmoudreza Babaei A Case - - PowerPoint PPT Presentation

▶
on microtargeting socially divisive ads
SMART_READER_LITE
LIVE PREVIEW

On Microtargeting Socially Divisive Ads: Mahmoudreza Babaei A Case - - PowerPoint PPT Presentation

Mar 13, 2024 357 likes •522 views

Filipe N. Ribeiro, Koustuv Saha, Mahmoudreza Babaei, Lucas Henrique, Johnnatan Messias, Fabricio Benevenuto, Oana Goga, Krishna P. Gummadi, and Elissa M. Redmiles. 2019. On Microtargeting Socially Divisive Ads: A Case Study of Russia- Linked Ad

slide-1
SLIDE 1

On Microtargeting Socially Divisive Ads: A Case Study of Russia-Linked Ad Campaigns on Facebook

Filipe N. Ribeiro* Koustuv Saha* Mahmoudreza Babaei Lucas Henrique Johnnatan Messias Fabricio Benevenuto Oana Goga Krishna P. Gummadi Elissa M. Redmiles

Filipe N. Ribeiro, Koustuv Saha, Mahmoudreza Babaei, Lucas Henrique, Johnnatan Messias, Fabricio Benevenuto, Oana Goga, Krishna P. Gummadi, and Elissa M. Redmiles. 2019. On Microtargeting Socially Divisive Ads: A Case Study of Russia- Linked Ad Campaigns on Facebook. In Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT* '19). ACM, New York, NY, USA, 140-149. DOI: https://doi.org/10.1145/3287560.3287580

slide-2
SLIDE 2

2016 U 2016 U.S. P Pres esiden dential al E Elec ections ns

2

slide-3
SLIDE 3

Soci

  • cial Media Adverti

rtising Soci

  • cial Media “Targ

rgeted” ” Adverti rtising

3

slide-4
SLIDE 4

These ads…

  • exploited Facebook’s targeted

advertising platform

  • mostly contained divisive or

polarizing content

# Views by topic (Source: USA Today)

slide-5
SLIDE 5
  • Facilitates selecting ad audience
  • These websites gather detailed demographic,

behavioral, location, and interest profiles of their users

  • Enables “microtargeting”: choose small

number of users with very specific attributes

Ta Targeted Ad Advertising

5

slide-6
SLIDE 6

Re Research Objectives

  • How divisive is the content of the IRA ads?
  • Were the targeting of the ads effectively done?
  • What features of the Facebook Ads platform were leveraged

in targeting the audience of the ads?

6

slide-7
SLIDE 7

US Election

101 102 103 104 105 106 Jun−2015 Jul−2015 Aug−2015 Sep−2015 Oct−2015 Nov−2015 Dec−2015 Jan−2016 Feb−2016 Mar−2016 Apr−2016 May−2016 Jun−2016 Jul−2016 Aug−2016 Sep−2016 Oct−2016 Nov−2016 Dec−2016 Jan−2017 Feb−2017 Mar−2017 Apr−2017 May−2017 Jun−2017 Jul−2017 Aug−2017

Amount

Impressions Clicks Cost (USD) Ads

  • 3,517 Ads
  • Each with corresponding cost, impressions, clicks
  • Avg.
  • Cost per Ad: 34.5 USD,
  • Impressions: 11,536,
  • Clicks: 1,062

10.8 1.6 1 0.8 0.56

5 10 IRA Retail Fitness Healthcare Finance

  • Avg. CTR (%) of Facebook Ads

DA DATA

7

slide-8
SLIDE 8

Hig High h Im Impac pact t Ads ds

  • Union of top 10% ads in

each of cost, clicks, impressions, and CTR

  • Run prior to the 2016

U.S. Elections

  • >80% of cost, clicks,

impressions

  • 485 such ads
slide-9
SLIDE 9

RQ1: Divisiveness of the IRA Ads

  • U.S census-representative surveys (40% liberals, 40%

conservatives, 20% moderate or neutral)

  • 15-30 responses per ad
  • Total 2,886 unique respondents

Reporting Approval False Claims

If they would report the ad shown Reaction: Approval / Disapproval If they could identify false claims

9

slide-10
SLIDE 10

Reporting

  • Whether they would

“report / flag the content”

  • f the ad shown.
  • If they would, why do they

find the ad inappropriate For over 73% of the ads, at least 20% respondents reported the ad.

5 10 15 20 25 30

% Reported Responses

It’s sexually inappropriate It’s violent or prohibited content It’s offensive It’s misleading or a scam I disagree with it It’s a false news story Its a spam Something else

Incites racism Creates divide in the society

slide-11
SLIDE 11

RQ1: Divisiveness of the IRA Ads

Reporting Approval False Claims

Ideological divisiveness

11

slide-12
SLIDE 12

Reported by the Liberals Not Reported by the Conservatives Reported by the Conservatives Not Reported by the Liberals

12

slide-13
SLIDE 13

RQ2: How well was the targeting done?

Reporting Approval False Claims

50 100 150 200 250 300 350 400

Ad

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Proportion Reported

Non-Targeted Targeted

50 100 150 200 250 300 350 400

Ad

−1.0 −0.5 0.0 0.5 1.0 1.5

Approval Score

Non-Targeted Targeted

50 100 150 200 250 300 350

Ad

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45

Proportion FCs

Non-Targeted Targeted

13

slide-14
SLIDE 14

RQ3: How the Facebook Ads Platform helped?

Attribute-based targeting

  • Demographics (gender, age, location, language)
  • Interests (eg. interested in African-American civil rights movement)
  • Behaviors (eg. Individuals with ethnic affinity: African-American)

Facebook attribute suggestion. (eg. similar pages with demographic skewness): 64% of the ads used this feature

14

slide-15
SLIDE 15
  • Characterized the Russia-linked ad campaign on

Facebook

  • These ads were ideologically divisive
  • Ten times more effective than a typical Facebook ad
  • Biased especially in race and political leaning
  • Ads targeted the users who are less likely to identify

the inappropriateness

  • Facebook Ads platform can likely be abused to create

social discord

Im Implic plicatio tions ns

15

slide-16
SLIDE 16

Thank You koustuv.saha@gatech.edu | koustuv.com

http://www.socially-divisive-ads.dcc.ufmg.br/

16

Filipe N. Ribeiro, Koustuv Saha, Mahmoudreza Babaei, Lucas Henrique, Johnnatan Messias, Fabricio Benevenuto, Oana Goga, Krishna P. Gummadi, and Elissa M. Redmiles. 2019. On Microtargeting Socially Divisive Ads: A Case Study of Russia-Linked Ad Campaigns on Facebook. In Proceedings of the Conference on Fairness, Accountability, and Transparency (FAT* '19). ACM, New York, NY, USA, 140-149. DOI: https://doi.org/10.1145/3287560.3287580