Table of links Umhlahlandlela ABSTRACT Ukuhlobisa INTRODUCTION Ukuhlolwa SEARCH ENGINES AND WOMEN’S DESCRIPTIVE REPRESENTATION I-search engines kanye ne-female representation MEASURING THE EXTENT OF ALGORITHMIC REPRESENTATION Ukulinganiswa kwe-extenze ye-algorithmic representation GENERAL DISCUSSION, CONCLUSION   AND   REFERENCES Ukubuyekezwa, Ukubuyekezwa Waze Imibuzo SEARCH ENGINES AND WOMEN’S DESCRIPTIVE REPRESENTATION I-search engines kanye ne-female representation I-Atkeson ne-Carrillo 2007; i-Bratton ne-Ray 2002). Ukukhishwa kwama-descriptive amaqembu amancane lihlanganisa iziphumo zopolitiki ezinzima izinzuzo zayo eziphambili, okunciphisa izixazululo zayo zendawo (i-Bratton ne-Ray 2002; i-Hessami ne-Fonseca 2020; i-Lowande et al. 2019; i-Wängnerud 2009; i-Mendelberg et al. 2014). Ukuqhathaniswa kwamakhemikhali kumasipala zihlanganisa iziphumo zomthetho (i-Stokes-Brown ne-Dolan 2010). Ukuba izivakashi zihlanganisa ngokufanele isakhiwo se-demographic yama-population yabo "ukuthumela isinyathelo ebalulekile Izifundo zihlanganisa i-media discourse njenge-factor ebalulekile esihlanganisa indlela ababhaliswe ababhaliswe futhi uma babona izinzuzo zabo njengoko zihlanganiswa (Kahn 1994). Umphumela obungapheliyo kulesi isikhungo se-policy under- and misrepresents minoritized groups (Gershon 2013; Van Der Pas and Aaldering 2020) futhi ukuthi lezi zihlanganisa ku-coverage kuholele ekunciphiseni ukubuyekeza ababhaliswe ababhaliswe ababhaliswe ababhaliswe (Rohrbach et al. 2023; van Oosten et al. 2024). Kulesi nqaku, thina siphinde sishintshele okufakelwa kakhulu kulesi isikhungo se-motivation futhi ibonise isikhungo se-algorithm. Algorithmic representation as outcome of structural inequalities I-Shehata ne-Strömbäck 2014; i-Kahn 1994). I-gender mediation thesis ivumela ukuthi ukuhanjiswa kwegama le-political kuya kwegama le-mediated iyatholakala ngokusebenzisa ama-filters ahlukahlukene ze-journalism (Gidengil ne-Everitt 2000; i-Trimble et al. 2022). Lezi zifiltrati zihlanganisa izakhiwo zokusebenza ze-gender ezivela kumazwe amabhizinisi, izinhlelo zokuthumela namabhizinisi kanye nezinsizakalo zayo zebhizinisi (i-Riedl et al. 2022). Lezi zofiltrated mediation zihlanganisa amabhizinisi nge-coverage eyenziwe ngamanani kanye nenkinga. Lezi zibonakalo zihlanganisa njengoba zihl Ngaphezu kwalokho, ku-AI-powered search enginesha-but also social media feed and news recommender systems-structure what accounts of politics voters get to see (and which they do not; bheka isib. Wallace 2018; Friesen et al. 2021). Ngaphezu kwalokho, ku-”post-broadcast democracies”, i-gate-keeping influence of journalists is a more and more confidence in the information obufumana nge-Internet searches than in traditional news media (bheka Stier et al. 2022). Izifundo zibonisa ukuthi izinjini zokufunda ezifana ne-Google ziye kwenziwa izindawo eziyinhloko zebhizinisi zebhizinisi zebhizinisi (Trevisan et al. 2018; Urman and Makhortykh 2023). Ngaphezu kwalokho, emzimbeni amanye, abantu I-H1: I-Based Line Bias: Abesifazane abalandeli kakhulu ku-Google image search (i-gender neutral) Uma umqondo we-masculinized political and social reality iveza ku-algorithmic underrepresentation of women on Google, then the extent of this underrepresentation should vary across societies with different political benchmarks regarding women's descriptive representation.4 Ngokusho ukususwa kwezinto zokuxhumana, i-Niven (1998, 2004) ibonise i-under/misrepresentation of women in political media coverage as a function of the scarcity of women leadership roles. Lokhu ukubuyekeza ku-models stereotype dynamic that emphasize that stereotypes are shaped by distribution of women and men among distinct occupational roles—that is, in politics, their presence in office (Diekman et al. 2004; Eagly et al. 2020; van der Pas et al. 2023). Ngokungafani nabantu I-H2: I-Distribution Bias: I-algorithmic underrepresentation ye-Women iyahambisana kahle ne-descriptive representation yayo emhlabeni wonke kodwa akuyona. Algorithmic representation as a driver of political perceptions Ngaphezu kwalokho, i-Shah al-direct iye yasungulwa i-hypothesis ukuthi izimpendulo ze-Chorock ziye zithunyelwe ngama-algorithmically engaphansi kuka-black politicians ku-Google output kungabangela i-labeling ye-race ye-individual ezivamile. Ngaphandle kokuphumelela kwezimpendulo ezine ze-3, izimpendulo ze-labeling zihlanganisa kuphela ku-gender futhi zihlanganisa ingxaki ze-corock futhi zihlanganisa izimo ze-race ngenxa yokuqala. Okokuqala, ukucubungula inani le-non-white politicians ku-Google output kungabangela ukucubungula izimpendulo ze-race ze-individual ezijwayelekile. Ngaphandle kokuphumelela kwezimpend Thola ukuthi umphumela we-algorithmic isebenza ngaphakathi kwesimo se-discrimination strategic ngezinyathelo ezimbili. Okokuqala, umphumela we-minoritized groups ku-search engine outputs ukuphazamiseka imibuzo ye-individual ye-descriptive representation of those groups (H3). Ngakho-ke sinikeza ukuthi umphumela we-search engine isebenza njenge-heuristic yobugcisa umphumela we-gender perception (Lau and Redlawsk 2001; McDermott 1997), okuyinto "ukudluliselwa okuqhubekayo kumadoda kanye nokudluliselwa kwe-female ku-selectivity yayo" (Ashworth et al. 2024, 290). Njengoba abaninzi abacwaningi ayikho umqondo oluthile lokuhlanganiswa kwe-demographic ephakeme kwez I-H3: I-Perceptual Bias: I-algorithmic underrepresentation ye-female ne-non-white politicians ivimbele ama-estimates ye-descriptive representation yabo. Okokuqala, lezi zibonakalayo ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingenalutho ezingen Ngaphandle kokunciphisa ukubuyekeza okuhlobene kubangelwa kubangelwa ku-minority groups, kukhona izizathu ukholelwa ukuthi ukunciphisa ukubuyekeza okuhlobene kuncike ukunciphisa ukufaneleka kwamabhayisikili, etholakala njenge-perception ukuthi isivumelwano asebenzayo imibuzo yabo (H4b; Atkeson and Carrillo 2007). Ngokuvamile, Stauffer (2021) kubonisa ngokucacileyo kwimeko yabasetyhini abacwaningi ukuthi kubangelwa kubangelwa kwamabhayisikili ku-inclusion yabasetyhini kuncike ngokucacileyo (bheka nangokuthi Dolan and Sanbonmatsu 2009). Sihlola le arguments futhi lithunyelwe kubangelwa abacwaningi non-white. I-H4: I-strategic bias: Ukubuyekezwa kwe-descriptive representations ngokuvamile kubuyekeza ukubuyekezwa kwamabhayisikili (a) ukubuyekezwa kwamabhayisikili nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane nabesifazane. 
 
 
 Umbhali : 
 
 
 
 U-Tobias Rohrbach Mykola Makhortykh I-Maryna Sydorova Umbhali : 
 
 
 
 U-Tobias Rohrbach Mykola Makhortykh I-Maryna Sydorova 
 
 Okuzenzakalelayo iyatholakala ku-archiv ngaphansi kwe-license ye-CC by 4.0 Deed (i-Attribution 4.0 International). Okuzenzakalelayo Ngaphandle kwe-CC by 4.0 Deed (i-Attribution 4.0 International) isicelo. available on arxiv I-Archive ye-Archive

Part of HackerNoon's growing list of open-source research papers, promoting free access to academic material.

Bias in Algorithmic Technology Exposed 

undefined's blog

Lo msindo ukhiqizwa ngolimi lokuqala lwendaba!

Ungayifaka kanjani i-search engama-gender gaps e-political representation

About Author

IMIBONO

HANG TAGS

LESI SIHLOKO SETHULWE NGAPHAKATHI

Related Stories

The Billionaires F*cked Up

Zulu Network: Moving The Bitcoin Economy Forward With a Two-Tiered Bitcoin Layer 2 Architecture

The Billionaires F*cked Up

Zulu Network: Moving The Bitcoin Economy Forward With a Two-Tiered Bitcoin Layer 2 Architecture

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps