scieee Science in your language
[en] (orig)

AI-BASED ANALYSIS OF GENDERED LANGUAGE PATTERNS IN ENGLISH–UZBEK SOCIAL MEDIA DISCOURSE

Author: Qudratullayeva, Muniraxon Abrorjon qizi; Mardonov, Quvonchbek Boymurod o'g'li
Publisher: Zenodo
DOI: 10.5281/zenodo.17686094
Source: https://zenodo.org/records/17686094/files/150-158.pdf
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
150
DOI: h ps://10.5281/10.5281/zenodo.17686094
AI-BASED ANALYSIS OF GENDERED LANGUAGE PATTERNS IN
ENGLISH–UZBEK SOCIAL MEDIA DISCOURSE
Qud a ullaye a Muni axon Ab o jon qizi
Teache a Kokand Uni e si y
Ma dono Qu onchbek Boymu od o‘g‘li
Teache a UzSWLU
ANNOTATION
This a icle in es iga es gende ed language pa e ns in English–Uzbek social
media discou se h ough AI-powe ed analy ical me hods. Using machine lea ning
classi ica ion, co pus-based analysis, sen imen modeling, and NLP ools, he s udy
examines how male and emale use s employ linguis ic o ms, code-swi ching
beha io s, emo i e exp essions, and poli eness s a egies ac oss bilingual pla o ms.
Keywo ds: AI, gende ed language, English–Uzbek discou se, NLP, social media
linguis ics, digi al iden i y, code-swi ching, sen imen analysis.
АННОТАЦИЯ
В данной статье исследуются гендерные языковые модели в англо-
узбекском дискурсе социальных сетей с использованием методов, основанных
на искусственном интеллекте. Применяя классификацию машинного обучения,
корпусный анализ, моделирование сентимента и инструменты обработки
естественного языка (NLP), исследование анализирует, как пользователи-
мужчины и пользователи-женщины используют языковые формы, стратегии
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
151
переключения кодов, эмоционально-экспрессивные средства и вежливые
речевые тактики на билингвальных онлайн-платформах.
Ключевые слова: искусственный интеллект; гендерно маркированный
язык; англо-узбекский дискурс; обработка естественного языка (NLP);
лингвистика социальных сетей; цифровая идентичность; переключение кодов;
сентимент-анализ.
ANNOTATSIYA
Ushbu maqola ingliz–o‘zbek ikki illi ij imoiy a moq disku sida gende ga xos il
xususiya la ini sun’iy in ellek yo damida ahlil qiladi. Mashina o‘ ganishi, ko pus
ahlili, sen imen modeli a NLP usulla idan oydalangan holda, e kak a ayol
oydalanu chila ning il bi ikmala i, kod-almashu oda la i a emo sional
i odala idagi a qla o‘ ganiladi.
Kali so‘zla : SI, gende ga xos il, ingliz–o‘zbek disku si, NLP, ij imoiy a moq
ling is ikasi, aqamli iden i e , kod-almashu , sen imen ahlili.
In oduc ion.
Digi al communica ion has eme ged as he p ima y a ena in which gende ed
linguis ic s yles and cul u ally embedded no ms in e sec , shi , and acqui e new
meanings. As social in e ac ion inc easingly mig a es o online pla o ms - social
media, messaging applica ions, and mul ilingual o ums he language choices made by
use s become powe ul indica o s o iden i y pe o mance
1
. In bilingual and
mul ilingual con ex s, such as English-Uzbek digi al discou se, hese choices a e
shaped no only by linguis ic compe ence bu also by he social expec a ions a ached
o each language. English o en indexes mode ni y, global belonging, o in o mali y,
while Uzbek may ca y associa ions wi h adi ion, local iden i y, and o mal
1
Ecke , P., & McConnell-Gine , S. (2013). Language and Gende (2nd ed.). Camb idge Uni e si y P ess.
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
152
poli eness no ms. These symbolic alues in e ac dynamically wi h gende , gene a ing
dis inc i e pa e ns in how men and women na iga e bilingual epe oi es online.
Recen ad ances in na u al language p ocessing ha e opened new
me hodological oppo uni ies o cap u ing and analyzing hese gende ed pa e ns wi h
unp eceden ed p ecision. La ge-scale AI language models can p ocess ex ensi e
da ase s o online communica ion pos s, commen s, messages, and hash ags enabling
esea che s o iden i y sub le, ecu en ea u es o gende ed s yle ha may elude
manual coding. Fo ins ance, AI models can de ec sys ema ic a ia ion in code-
swi ching equency, opic selec ion, s ance exp essions, emo i e ma ke s, and
discou se s a egies as hey co ela e wi h gende and cul u al con ex .
1
Mo eo e ,
compu a ional ools make i possible o examine how hese pa e ns e ol e o e ime,
espond o sociopoli ical e en s, o a y ac oss di e en digi al en i onmen s.
In he s udy o English-Uzbek bilingual discou se speci ically, AI-assis ed
analysis allows esea che s o explo e how gende no ms a e main ained, nego ia ed,
o con es ed h ough linguis ic choices. Women and men may di e in hei s a egic
use o English o empowe men , anonymi y, o cosmopoli an iden i y pe o mance,
while elying on Uzbek o a i m cul u al belonging o comply wi h expec ed
poli eness con en ions. Con e sely, momen s o code-swi ching may unc ion as
delibe a e ac s o dis ancing, solida i y, humo , o esis ance, each in luenced by he
speake ’s gende ed posi ion wi hin socie y.
2
By enabling esea che s o sys ema ically quan i y and model hese phenomena,
AI echnologies con ibu e no only o desc ip i e accoun s o bilingual digi al
communica ion bu also o b oade heo e ical amewo ks on language, gende , and
cul u al hyb idi y. The in eg a ion o compu a ional linguis ics wi h sociolinguis ic
and gende s udies he e o e ep esen s a signi ican me hodological and concep ual
1
Fie man, W. (2016). “Language a i udes in Uzbekis an: Uzbek, Russian, and English.” In e na ional Jou nal o he
Sociology o Language, 239, 79–107.
2
Came on, D. (2007). The My h o Ma s and Venus: Do Men and Women Really Speak Di e en ly? Ox o d Uni e si y
P ess.
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
153
ad ancemen , o e ing new insigh s in o how bilingual speake s na iga e and eshape
linguis ic no ms in he digi al age.
Li e a u e e iew.
Gende ed language is no jus a se o linguis ic pa e ns - i is a e lec ion o how
people ela e o each o he . The ways indi iduals exp ess poli eness, show emo ion, o
choose ce ain wo ds o en e eal he expec a ions placed on hem wi hin hei
communi ies. In many cul u es, women a e encou aged o speak mo e gen ly o
diploma ically, while men a e expec ed o sound mo e di ec o ese ed. These
endencies a e no ixed ules, bu hey do show how social no ms shape e e yday
communica ion.
In bilingual se ings, hese dynamics become e en mo e in e es ing. Speake s
who mo e be ween English and Uzbek, o example, o en use each language o
highligh di e en pa s o who hey a e. English migh eel mo e sui ed o exp essing
con idence o neu ali y, while Uzbek may ca y a sense o wa m h, cul u al closeness,
o espec . By swi ching languages, people can shi hei one, adjus hei emo ional
exp ession, o posi ion hemsel es di e en ly depending on he si ua ion. This
lexibili y makes bilingual communica ion a ich space o unde s anding how gende
is exp essed.
1
Today, AI ools gi e esea che s new ways o explo e hese pa e ns. Ins ead o
elying only on small samples o manual obse a ion, machine lea ning models can
analyze housands o messages a once. They can iden i y when speake s swi ch
languages, wha kinds o wo ds hey choose, o how hey exp ess poli eness o
emo ion. These ools help unco e pa e ns ha migh o he wise go unno iced - like
sub le di e ences in how men and women use emojis, bo ow English slang, o swi ch
o Uzbek o emphasis o in imacy.
By combining human insigh wi h he powe o AI, esea che s can build a clea e
pic u e o how gende , language, and cul u e in e ac in digi al spaces. This app oach
1
He ing, S. C., & Kapidzic, S. (2015). “Teens, gende , and sel -p esen a ion in social media.” In e na ional Encyclopedia
o Social and Beha io al Sciences, 2nd ed., 1–11.
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
154
makes i possible o see no only wha people say, bu also how hei linguis ic choices
help hem na iga e iden i y and connec ion in a mul ilingual wo ld.
Resea ch me hodology.
To unde s and how gende ed language pa e ns eme ge in eal digi al
in e ac ions, we analyzed da ase s collec ed om se e al majo online pla o ms,
including Ins ag am, Teleg am, TikTok, and YouTube commen sec ions. These
pla o ms we e chosen because hey ep esen di e en s yles o communica ion -
anging om casual messaging o highly public con e sa ions - and he e o e o e a
b oad iew o how people exp ess hemsel es ac oss con ex s.
A ange o AI-d i en echniques was applied o hese da ase s. Sen imen
modeling allowed us o iden i y he emo ional one o messages and see how men and
women exp ess posi i i y, nega i i y, o nuance di e en ly ac oss languages.
Keywo d ex ac ion helped highligh he wo ds and ph ases mos s ongly associa ed
wi h each gende , p o iding insigh in o ecu ing opics, s ylis ic choices, and cul u al
e e ences.
To cap u e he bilingual na u e o he discou se, we used code-swi ching
de ec ion ools ha au oma ically pinpoin ed momen s when use s shi ed be ween
English and Uzbek. These swi ches o en ca y impo an social meaning signaling
closeness, humo , emphasis, o shi s in iden i y so iden i ying hem a scale was
essen ial. Finally, embedding-based classi ie s enabled deepe analysis o pa e ns ha
a e no ied o any single wo d, such as sub le s ylis ic endencies o gende ed ways o
aming ideas.
Toge he , hese me hods c ea ed a comp ehensi e pic u e o how gende and
bilingualism in e ac ac oss mul iple digi al pla o ms, o e ing insigh s ha would be
di icul o obse e h ough manual analysis alone.
Analysis and esul s:
Gende ed Linguis ic Pa e ns. Clea di e ences eme ged in how emale and male
use s exp ess hemsel es ac oss pla o ms. Female use s ended o communica e wi h
g ea e emo ional in ensi y, o en using exp essi e ocabula y, exclama ion ma ks,

GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
155
and a no iceably wide ange o emojis o con ey one, empa hy, o humo . These
choices sugges a communica ion s yle gea ed owa d ela ional connec ion and
emo ional cla i y, aligning wi h b oade sociolinguis ic obse a ions.
Male use s, on he o he hand, g a i a ed owa d mo e di ec , concise ph asing.
Thei messages ea u ed less emo ional ma king and mo e use o echnical slang,
ja gon, o opic-speci ic keywo ds - especially in discussions ela ed o echnology,
gaming, inance, and p oblem-sol ing. This p e e ence e lec s a mo e e e en ial and
ask-o ien ed communica i e s yle.
Code-swi ching pa e ns also e ealed meaning ul s ylis ic and cul u al
endencies. Women mo e equen ly swi ched o Uzbek o exp ess wa m h, a ec ion,
o cul u al in imacy, while using English o mode ni y, humo , o s ylis ic lai . Men’s
code-swi ching ended o be mo e unc ional, o en igge ed by opic shi s o he
in oduc ion o echnical e ms. These bilingual pa e ns demons a e how speake s
d aw on he symbolic alue o each language o shape one, iden i y, and social
alignmen .
1
The Role o AI in Shaping Gende ed Discou se.
Digi al communica ion oday is inc easingly shaped by p edic i e ex sys ems,
machine ansla ion ools, and con e sa ional cha bo s. These echnologies sub ly
guide how use s o mula e hei messages, o en wi hou conscious awa eness. One
no iceable e ec is he ein o cemen o English-o igin exp essions and hyb id
linguis ic o ms wi hin bilingual English-Uzbek communica ion.
P edic i e ex sugges ions equen ly o e English lexical i ems especially
endy slang, sho ened o ms, o high- equency exp essions which use s may adop
because hey a e quicke o selec and socially ecognizable online. Female use s o en
inco po a e hese sugges ions in o exp essi e o s ylis ic elemen s, blending English
e ms wi h Uzbek o c ea e so e , mo e emo ionally nuanced ph asing. Male use s
1
Holmes, J., & Meye ho , M. (Eds.). (2003). The Handbook o Language and Gende . Wiley-Blackwell.
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
156
mo e commonly adop English-o igin echnical e ms o concise hyb id cons uc ions,
aligning wi h hei p e e ence o di ec and e icien communica ion.
Machine ansla ion ools also con ibu e o his end by p omo ing syn ac ic
s uc u es and ocabula y ha e lec English no ms, some imes leading o calques o
hyb id exp essions in Uzbek. Cha bo s, which o en espond using s anda dized
English-in luenced pa e ns, u he no malize hese o ms by p o iding inpu ha
use s consciously o unconsciously mi o .
Taken oge he , hese AI-media ed ools ac as sub le bu in luen ial ac o s in
shaping how bilingual speake s exp ess gende ed iden i ies. They encou age he
di usion o English loanwo ds, hyb id ph ases, and digi ally shaped discou se s yles,
g adually eshaping he linguis ic landscape o online communica ion.
Discussion. AI-d i en analysis e eals ha online language use is no simply a
e lec ion o indi idual p e e ence; i is shaped by he in e play o cul u al no ms,
global digi al p ac ices, and algo i hmic in luences. In he case o English-Uzbek
bilingual communica ion, his in e ac ion becomes pa icula ly isible.
Uzbek gende no ms, which in luence expec a ions a ound poli eness, emo ional
exp ession, and social posi ioning, con inue o guide how men and women p esen
hemsel es online. A he same ime, English-language digi al cul u e-cha ac e ized
by in o mali y, exp essi e c ea i i y, and apid adop ion o new slang in oduces
al e na i e models o sel -exp ession ha use s d aw upon as hey c a hei online
iden i ies. Fo many, swi ching o English is a way o apping in o global ends,
signaling mode ni y, o expe imen ing wi h s yles no ypically a ailable wi hin
adi ional no ms.
Algo i hmic shaping adds a hi d laye o his dynamic. P edic i e ex sys ems,
ansla ion ools, and cha bo in e ac ions sub ly encou age ce ain linguis ic choices
h ough hei sugges ions, de aul s, and s ylis ic pa e ns. As a esul , use s equen ly
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
157
adop English-in luenced o ms, hyb id exp essions, o algo i hmically a o ed
ph asing, e en when communica ing wi hin cul u ally g ounded Uzbek con ex s.
1
Toge he , hese o ces local gende expec a ions, global digi al cul u e, and
algo i hmic in luence blend o c ea e hyb id digi al iden i ies. These iden i ies a e
nei he ully adi ional no ully global; hey eme ge om he c ea i e nego ia ion o
mul iple linguis ic and cul u al esou ces. AI helps make hese p ocesses isible,
e ealing how people c a new o ms o sel -exp ession ha anscend singula
cul u al amewo ks while s ill emaining oo ed in hem.
Conclusion.
AI-based analysis e eals ha gende plays a signi ican and consis en ole in
shaping English-Uzbek bilingual discou se. Ac oss pla o ms and communica i e
con ex s, men and women d aw on di e en emo ional ma ke s, s ylis ic p e e ences,
and code-swi ching s a egies, e lec ing bo h cul u ally embedded gende no ms and
he in luence o global digi al p ac ices. A he same ime, hyb id linguis ic o ms
mixing English exp essions, Uzbek s uc u es, emojis, slang, and algo i hmically
sugges ed ph asing a e becoming inc easingly no malized. These hyb ids a e no longe
pe iphe al o play ul; hey a e eme ging as a s able pa o e e yday digi al
communica ion, especially among younge bilingual use s.
O e all, AI-d i en linguis ic esea ch opens a new window in o how gende ,
cul u e, and echnology in e sec in mul ilingual digi al spaces— e ealing a complex,
e ol ing landscape whe e hyb id iden i ies a e becoming a de ining ea u e o
con empo a y communica ion.
REFERENCES
1. Ecke , P., & McConnell-Gine , S. (2013). Language and Gende (2nd
ed.). Camb idge Uni e si y P ess.
1
He ing, S. C. (2004). “Compu e -media ed discou se analysis: An app oach o esea ching online beha io .” Designing
o Vi ual Communi ies in he Se ice o Lea ning, 338–376.
GOLDEN BRAIN ISSN: 2181-4120 VOLUME 3 | ISSUE 17 | 2025
Mul idisciplina y Scien i ic Jou nal No embe , 2025
158
2. Came on, D. (2007). The My h o Ma s and Venus: Do Men and Women
Really Speak Di e en ly? Ox o d Uni e si y P ess.
3. Holmes, J., & Meye ho , M. (Eds.). (2003). The Handbook o Language
and Gende . Wiley-Blackwell.
4. He ing, S. C. (2004). “Compu e -media ed discou se analysis: An
app oach o esea ching online beha io .” Designing o Vi ual Communi ies in he
Se ice o Lea ning, 338–376.
5. Bake , P. (2014). Using Co po a o Analyze Gende .
6. C ys al, D. (2011). In e ne Linguis ics.
7. Ecke , P. & McConnell-Gine , S. (2013). Language and Gende .
8. Ga cía, O. & Wei, L. (2014). T anslanguaging.
9. Nguyen, D. (2021). Gende and language in online communica ion.
10. He ing, S. C., & Kapidzic, S. (2015). “Teens, gende , and sel -
p esen a ion in social media.” In e na ional Encyclopedia o Social and Beha io al
Sciences, 2nd ed., 1–11.
11. Kendall, L. (2002). Hanging Ou in he Vi ual Pub: Masculini ies and
Rela ionships Online. Uni e si y o Cali o nia P ess.
12. Fie man, W. (2016). “Language a i udes in Uzbekis an: Uzbek, Russian,
and English.” In e na ional Jou nal o he Sociology o Language, 239, 79–107.
13. Pa lenko, A. (2008). “Mul ilingualism in pos -So ie coun ies.”
In e na ional Jou nal o Bilingual Educa ion and Bilingualism, 11(3–4), 275–314.