Management and Future Technologies | Volume 1, Issue 2 | № 002

June, 2024 № 002 Volume 1, Issue 2 Management and Future Technologies SCIENTIFIC JOURNAL journal.umft.uz

Jurnal sohalari: menejment, dasturiy injiniring, sun’iy intellekt texnologiyalari, kompyuter injiniringi, infokommunikatsiya injiniringi, axborot xavfsizligi, raqamli iqtisodiyot, pedagogika va psixologiya, matematika va fizika Tahririyat kengashi raisi Muharrirlar t.f.d., prof. O‘tkir Xamdamov “University of Management and Future Technologies” universiteti rektori Farhod Ahmedov (J. Koreya) Sherzod Mustafaqulov (O‘zbekiston) Akmalbek Abdusalomov (J. Koreya) Bahtiyor Akmuradov (O‘zbekiston) Sherzod Abdullayev (O‘zbekiston) Shabir Ahmad (J. Koreya) Dilshod Rahmatov (Germaniya) O.A. Hidayov (Germaniya) Akmaljon Abdullayev (O‘zbekiston) Jinsoo Cho (J. Koreya) Jamshid Elov (O‘zbekiston) Young Im Cho (J. Koreya) Fazliddin Maxmudov (J. Koreya) Alpamis Kutlimuratov (O‘zbekiston) Faheem Khan (J. Koreya) Lilia Tightiz (J. Koreya) Shahnoza Sultanova (O‘zbekiston) Safiullah Khan (Buyuk Britaniya) Seytkamal Medetov (Fransiya) Avaz Qaxxorov (O‘zbekiston) Doston Xasanov (O‘zbekiston) Sabina Umirzakova (J. Koreya) Elmurod Urinov (O‘zbekiston) Muhriddin Umarov (O‘zbekiston) Urmanov Odil (J. Koreya) Sobir Radjabov (O‘zbekiston) Bosh muharrir t.f.f.d. dot. Muhriddin Muxiddinov “University of Management and Future Technologies” universiteti Ilmiy va uslubiy ishlar bo‘yicha prorektori Tahririyat kengashi a’zolari t.f.d., prof. Xakim Zaynidinov t.f.d., prof. Muxammadjon Musayev t.f.d., prof. Shavkat Fozilov f-m.f.d. prof. Aripov Mersaid f-m.f.d. prof. Аloev Raxmatillo t.f.d., prof. Marat Raxmatullayev t.f.d., prof. Dilnoz Muxamediyeva i.f.d., prof. Toxir Hasanov i.f.d., prof. To‘lqin Boboqulov t.f.d., prof. Jamshid Sultonov f-m.f.d. prof. Dildora Muhamediyeva i.f.d., prof. Sherzod Rajabov p.f.d., prof. Qurbonniyoz Panjiyev p.f.d., prof. Muhabbat Hakimova t.f.d., prof. Nargiza Usmanova p.f.d., prof. Nishonboy Kiyamov p.f.d., prof. Qutlug‘bek Qodirov t.f.d., dot. Halimjon Xujamatov t.f.d., dot. Ibragim Atadjanov Dizayner Eshmetov Daler Jurnal O‘zbekiston Respublikasi Prezidenti Administratsiyasi huzuridagi Axborot va ommaviy kommunikatsiyalar agentligidan 212842-raqamli guvohnoma bilan 26.01.2024 sanasida ro‘yxatdan o‘tkazilgan.

MUNDARIJA TEXNIKA FANLARI TILNI QAYTA ISHLASHDA QO‘LLANILADIGAN NLP MODELLARI 6 TABIIY TAHLILI Djamshid Sultanov, Xusniya Axmedova REVOLUTIONIZING CUSTOMER SERVICE IN UZBEKISTAN 12 UZASSISTANT: WITH AI Abdinabi Mukhamadiyev, Jinsoo Cho DIAGNOSIS OF CATTLE DISEASES USING FUZZY LOGIC 17 ENHANCED ALGORITHMS Dilmurod Turimov USE OF INNOVATIVE AND DIGITAL TECHNOLOGIES IN TEACHING 22 THE PROGRAMMABLE DIGITAL DEVICES Xalima Abasxanova, E’zoza Musoqulova and Sabrina Sotvoldiyeva 28 VAZIYATGA SEZGIR AQLLI MUHITNING NAMUNAVIY ARXITEKTURASI Nargiza Usmanova, Zafarjon G‘ayratov 39 A COMPARATIVE ANALYSIS OF TRADITIONAL AND DEEP LEARNING BASED APPROACHES FOR FACE DETECTION: PERFORMANCE, ADVANCEMENTS, AND CHALLENGES Elmurod Urinov, Nurbek Kholmurodov 44 GISTOLOGIK TASVIRLARDAGI MAVJUD SOHALARNI O‘LCHASH USULI Meliyev Farxod, Nugmanova Mavluda

O‘QITISH ALGORITMLARI ASOSIDA KO‘ZI OJIZ INSONLAR UCHUN 52 CHUQUR AQLLI BOSH KIYIM ISHLAB CHIQISH Muhriddin Umarov, Farrux Axmedov, Jasurbek Djuraev IQTISODIY O‘SISH SHAROITIDA AXBOROT-KOMMUNIKATSIYA 62 BARQAROR TEXNOLOGIYALARINING O‘RNI Shaxlo Xo‘jamatova AVTOMOBILLARINING DAVLAT RAQAMINI TANIB OLISH 70 O‘ZBEKISTON ALGORITMI ISHLAB CHIQISH Raxmatilla G‘aybullayev TEXNOLOGIYALAR: ASOSIY TUSHUNCHALAR, ZAMONAVIY 78 RAQAMLI DUNYOGA TA’SIRI VA MISOLLAR Zafarjon G’ayratov, Mirjalol Najmiyev va Sevinch Umurzokova IJTIMOIY FANLAR OF THE PROCESS OF DIGITAL TRANSFORMATION IN SMALL 84 ANALYSIS AND MEDIUM-SIZED ENTERPRISES Yorqinjon Khusniddinov 90 MODELS AND METHODOLOGIES SOFTWARE DEVELOPMENT Muzaffarjon Abdulakhatov XO‘JALIGIDA RAQAMLASHTIRISHNI RIVOJLANTIRISHNING 96 QISHLOQ DASTLABKI SHARTLARI VA DIAGNOSTIKASI Alisher Burxanov, Samandar Berdiqulov, Inomjon Xabibullaev va Dilnovoz Azimova IQTISODIYOTNI RIVOJLANTIRISHDA KRIPTAVALYUTA 105 RAQAMLI BOZORINING O‘RNI Umidjon Oybekov, Bahromjon Boliyev

THE METHODOLOGY FOR FIXED ASSETS VALUATION IN THE 112 DEVELOPING CONTEXT OF A DIGITAL ECONOMY Odiljon Rikhsimbaev MATEMATIKA VA FIZIKA FANLARI ПОТЕРИ ВОДЫ ОТ ИЗМЕНЕНИЯ КОЭФФИЦИЕНТА 117 ЗАВИСИМОТЬ ПРОНИЦАЕМОСТИ ДНА И БОКОВЫХ СТЕНОК КАНАЛОВ Алишер Усмонов INNOVATSION YONDOSHUVLAR ORQALI FIZIKA DARSLARINI 128 YANGI SAMARALI TASHKIL ETISH Yakub Axmedov, Xasan Karimov OF GENERATING FUNCTIONS IN PROBABILITY AND 135 APPLICATION COMBINATORICS Munisa Ismoilova, Shakhlo Khabibullaeva, Zuhriddin Nazarov ИСПОЛЬЗОВАНИЯ ПЕРОВСКИТНЫХ ТАНДЕМНЫХ 142 ПЕРСПЕКТИВЫ СОЛНЕЧНЫХ ЭЛЕМЕНТОВ Нодирбек Ахмеджонов S SIMPLEKSDA VOLTERRA TIPIDAGI, UMUMIY HOLATDA BO‘LMAGAN 148 BA’ZI 4-DARAJALI STOXASTIK OPERATORLAR 2 Surojiddin Pardabayev

Volume 1, Issue 2 June 2024 6 Mualliflar Djamshid Sultanov Infokommunikatsiyalar injiniringi kafedrasi, Muhammad al-Xorazmiy nomidagi Toshkent axborot texnologiyalar universiteti, 100200, Toshkent, O‘zbekiston; sdjamshid@gmail.com Xusniya Axmedova Infokommunikatsiyalar injiniringi kafedrasi, Muhammad al-Xorazmiy nomidagi Toshkent axborot texnologiyalar universiteti, 100200, Toshkent, O‘zbekiston; h.ahmedova86@mail.ru *Mas’ul: sdjamshid@gmail.com Management and Future Technologies journal.umft.uz TABIIY TILNI QAYTA ISHLASHDA QO‘LLANILADIGAN NLP MODELLARI TAHLILI A nnotatsiya: So‘nggi yillarda Sun’iy intellekt (SI) texnologiyalari yuqori darajada rivojlandi va turli sohalarda, jumladan, banklarda, elektron pochta boshqaruvida, tibbiyotda va boshqa ko‘plab sohalarda qo‘llanilmoqda. NLP (Natural Language Processing - Tabiiy Tilni Qayta Ishlash) so‘zida ko‘plab modellarni o‘z ichiga oladi. Bu modellar, matnni tahlil qilish, tarjima qilish, dialog tizimi, mazmuni tahlil qilish va boshqa ko‘plab vazifalarni bajarish uchun ishlatiladi. Bu modellar katta matnlar va ko‘p maqsadli vazifalarni bajarishda muvaffaqiyatli natijalarga erishish uchun yaratilgan. Har bir modellarning o‘z xususiyatlari va foydalanish sohalari mavjud. Ushbu maqolada shunday modellarning bir nechtasining imkoniyatlari va o‘ziga xos xususiyatlari bo‘yicha ma’lumotlar keltirilgan. Kalit so‘zlar: tabiiy til, matn, NLP, RNN, Word2vec, GloVe, FastText. Copyright: © 2024 by the authors. Ushbu maqola Creative Commons Attribution (CC BY) litsenziyasi shartlari asosida tarqatiladigan ochiq foydalanish maqolasi hisoblanadi (https://creativecommons.org/ licenses/by/4.0/).

Management and Future Technologies journal.umft.uz ANALYSIS OF NLP MODELS USED IN NATURAL LANGUAGE PROCESSING A bstract: In recent years, Artificial Intelligence (AI) technologies have advanced greatly and are being used in various fields, including banking, email management, medicine, and many others. The term NLP (Natural Language Processing) includes many models. These models are used for text analysis, translation, dialog system, content analysis and many other tasks. These models are designed for successful results on large texts and multi-purpose tasks. Each model has its own characteristics and areas of use. This article provides information on the capabilities and characteristics of several such models. Keywords: natural language, text, NLP, RNN, Word2vec, GloVe, FastText. June 2024 Volume 1, Issue 2 7 Authors Djamshid Sultanov Department of Information Communications Engineering, Tashkent University of Information Technologies named after Muhammad al-Khwarazmi, 100084, Tashkent, Uzbekistan; sdjamshid@gmail.com Husniya Akhmedova Department of Information Communications Engineering, Tashkent University of Information Technologies named after Muhammad al-Khwarazmi, 100084, Tashkent, Uzbekistan; h.ahmedova86@mail.ru *Correspondence: sdjamshid@gmail. com Copyright: © 2024 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/ by/4.0/).

Volume 1, Issue 2 June 2024 Management and Future Technologies journal.umft.uz Kirish 8 Tilni modellashtirishning vazifalaridan biri oldingi matnni bilish asosida keyingi so‘zni bashorat qilishdir. Bu matn terish xatolarini tuzatish, avtoto‘ldirish, chatbotlar va boshqalar Ǥ juda ko‘p tarqoq ma’lumotlarni topishingiz mumkin. Tilni modellashtirishda alohida so‘zlar va so‘zlar guruhlari vektorlarga, ya’ni semantik munosabatlarni saqlaydigan ma’lum sonli tasvirlarga ko‘rsatiladi. So‘zning siqilgan vektor ko‘rinishi embedding deb ataladi. Bunday model oldingi n ta so‘zning kirish vektor ko‘rinishini oladi va jumlaning semantikasini "tushunish"ga harakat qiladi. Model treningi uzluksiz so‘zlar algoritmiga asoslangan. Kontekstli (qo‘shni) so‘zlar markaziy so‘zni bashorat qiladigan Ǥ Asosiy qism A bag of words (so‘zlar sumkasi) bu matnni vektor (so‘zlar to‘plami) sifatida ifodalash uchun qo‘llaniladigan modellardan biridir. Bunda matndagi har bir so‘z uning paydo bo‘lish soni bilan bog‘liq Ǥ Recurrent neural network (RNN) Elementlar orasidagi yo‘naltirilgan aloqalarga ega neyron tarmoqlar. Neyronning chiqishi kirishga qaytarilishi mumkin. Ushbu tuzilma sizga o‘ziga xos "xotira" ga ega bo‘lish va ma’lumotlar ketmaǦǡǡȏ͵ȐǤ Hammamizga ma’lum Google ko‘pgina tillar uchun oldindan o‘qitilgan ochiq kodli modellardan foydalanadi. Model inglizcha Google News 200B korpusida o‘qitilgan va 128 o‘lchovli o‘rnatishni hosil qiluvchi, oldinga uzatiladigan neyron tarmog‘ining uchta yashirin Ǥ ǣ - Oddiylik. Model tezda o‘rgatadi va o‘rnatishlarni yaratadi, bu ko‘pchilik oddiy ilovalar Ǥ - Ko‘plab tillarda oldindan tayyorlangan versiyalar mavjud. ǣ - Uzoq muddatli bog‘lanishlarni hisobga olmaydi. - Ǥ - Ǥ Word2vec 2013 yilda Google kompaniyasidan Tomas Mikolov so‘zlarning vektor ko‘rinishini o‘rganish uchun yanada samarali modelni taklif qildi Ǧ Word2vec. Usul ko‘pincha bir xil kontekstda uchraydigan so‘zlar o‘xshash ma’noga Ǥ til modellarini ishlab chiqishda keskin burilish yasadi. Word2Vec modeli uzluksiz so‘zlarni to‘plash algoritmi o‘rniga SkipǦ Ǥ Ǧmarkaziy so‘zlarga asoslanib, atrofdagi so‘zlarni bashorat qilish. Ǧgram modeli, bir so‘zning boshqa so‘zlar bilan hamkorlik qilishi mumkinligini aniqlash uchun so‘zlar orasidagi aloqalar tahlilini bajaradi. Bu model uchun kiritish qatorining uchinchi jihati so‘zlar va ularning konteksti bo‘lishi mumkin bo‘lgan qatorlarning ko‘p ma’lumotlaridan foydalaniladi. Bu ma’lumotlar bilan model o‘rganadi va so‘zlarning xususiyatlari (embedding) ni yaratadi. Natijada, har bir so‘z uchun bir qator kontekst so‘zlar da joylashishi va ularning ma’nolari bo‘yicha o‘rganiladiȏʹȐǤǦǡ

Management and Future Technologies journal.umft.uz June 2024 Volume 1, Issue 2 so‘zlarning vektor koordinatalarini olish orqali amaliyotlarda o‘zlashtirish imkonini beradi. Bu vektorlar, so‘zlarning aynan ularning kontekstida uchun ma’nolarga yaqinlik [1]. Ma’lumotlar tahlili, matematik va tarqalgan texnologiyalar yordamida bu vektorlar ustida amalga oshiriladi. Bu vektorlar, ko‘p yo‘nalishli ma’lumotlar tahlili, matematik va yadroli ustama ma’lumotlar tahlili (NLP) uchun muǤ - ǣǡͳǡͳǡͳ Ǥ - Model tezda o‘rganadi va o‘rnatishlarni yaratadi (hatto o‘zingiz ham). - O‘rnatishlar ma’noga ega, bahsli fikrlarni hal qilish mumkin. - Metodologiya boshqa ko‘plabȋǡʹ ȌǤ - So‘z darajasida o‘rganish: so‘z ishlatilgan jumla yoki kontekst haqida hech qanday ma’lumot yo‘q. - Birgalikda yuzaga kelishi e’tiborga olinmaydi. Model so‘zning qo‘llanish qarab turli xil ma’nolarga ega bo‘lishi mumkinligini hisobga olmaydi. Bu GloVeǦ ʹ Ǧdan afzal ko‘rishining asosiy sababidir. - Noma’lum yoki kam uchraydigan so‘zlarni to‘liq qayta ishlamaydi. GloVe (Global Vectors) Word2Vec bilan chambarchas bog‘liq: algoritmlar bir vaqtning o‘zida paydo bo‘lgan va so‘z vektorlarining talqin qilinishiga tayanadi. GloVe modeli hit statistikasidan samarali foydalanish muammosini hal qilishga harakat qiladi. GloVe so‘z vektorlarining ma stoxastik gradient tushishidan foydalangan holda ularning birgalikda paydo bo‘lish ehtimoli logarifmi o‘rtasidagi farqni minimallashtiradi. Olingan tasvirlar so‘zlarning vektor fazosining ǣyoraning turli sun’iy yo‘ldoshlarini yoki shaharning pochta indeksini uning nomi bilan bog‘lash mumkin[5]. ʹ Ǧda so‘zlarning birgalikda paydo bo‘lish chastotasi unchalik muhim emas, bu faqat qo‘shimcha o‘quv namunalarini yaratishga yordam beradi. Glo ma’lumotlarga tayanishdan ko‘ra, birgalikdagi hodisalarni hisobga oladi. So‘z vektorlari global o‘xshashlik asosida guruhlanadi. - Neyron tarmog‘isiz oddiy arxitektura. - Model tez va oddiy ilovalar uchun yetarli bo‘liǤ - GloVe Word2Vec imkoniyatlarini to‘ldiradi. U so‘z chastotasini qo‘shadi va ko‘pgina ʹ ǦǤ - Birgalikda sodir bo‘lish matritsasi global ma’lumotni taqdim etsaǦda, GloVe so‘z darajasida o‘qitilgan bo‘libqoladi va jumla va so‘z ishlatiladigan kontekst haqida kam ma’lumot beradi. - Noma’lum va kam uchraydigan so‘zlarni sifatsiz qayta ishlash. FastText adam bo‘ldi. Uni ishlab chiqishda bizga Word2VecǦdan tanish bo‘lgan Tomas Mikolov ishtirok etdi. So‘zlarni vektorlashtirish uchun bir vaqtning o‘zida skipǦǡ ȏͶȐǤ ʹ Ǧgramm modeli qo‘shildi. Har bir so‘z ma’lum 9

Volume 1, Issue 2 10 June 2024 Management and Future Technologies journal.umft.uz Ǧ Ǥ Klassifikator natijalari kam uchraydigan so‘zlar uchun juda mos keladi, chunki ular nǦ bo‘lingan. Word2Vec Ǧdan farqli o‘laroq, model noma’lum so‘zlar uchun qo‘shimchalar Ǥ - Nisbatan oddiy arxitektura: oldinga o‘tish, 1 kirish, 1 yashirin qatlam, bitta chiqish - Ǧgrammlar tufayli u kam uchraydigan va eskirgan so‘zlarǤ - So‘z darajasida o‘rganish: so‘z ishlatilgan jumla yoki kontekst haqida hech qanday ma’lumot bo‘lmaydi. - Birgalikda paydo bo‘lish e’tiborga olinmaydi, ya’ni model turli kontekstlarda so‘zning turli ma’nolarini hisobga olmaydiǤ Xulosa ǡ sohasida yangiliklar doimiy ravishda paydo bo‘lishi mumkin. Mavjud modellar o‘zlarini yanada rivojlantirish uchun ko‘plab sohalarga qo‘llanilmoqda va yangi modellar ham paydo bo‘lmoqda. ǡ Ǧǡǡǡ ǦE so‘z va tasvirlar kombinatsiyasini yaratishda muvaffaqiyat qozonmoqda. Bunday ǡ Ǥ Shuningdek, ko‘p kompaniyalar xizmatlarida NLP ni qo‘llashga qodir modellar ishlab Ǥǡǡ Ǥ ǡ o‘konlarda xizmat ko‘rsatish, ma’lumotlarni topish va foydalanuvchilarga yordam berishda keng qo‘llanilmoqda. NLP yangiliklarni ko‘rsatish bilan shug‘ullanmoqda. Ushbu soha, kiber xavfsizlik, tibbiyot, ta’lim, avtomatizatsiya va boshqa sohalar uchun e’tiboringizni jalb etuvchi yuqori potensialga ega.

Management and Future Technologies journal.umft.uz June 2024 Volume 1, Issue 2 Adabiyotlar 1. S. A. Majali, “The impact of empowering people with special needs from information and communication technology,” Arab Journal of Disability and Talent Sciences, vol. 4, no. 14, 2020. 2. M. R. Morris, A. Zolyomi, C. Yao, S. Bahram, J. P. Bigham, and S. K. Kane, “With most of it being pictures now, I rarely use it: Understanding Twitter’s evolving accessibility to blind users,” in Proc. the 2016 CHI Conference on Human Factors in Computing Systems, San Jose California, USA, 2016. 3. C. Morrison, E. Cutrell, A. Dhareshwar, K. Doherty, A. Thieme, and A. S. Taylor, “Imagining artificial intelligence applications with people with visual disabilities using tactile ideation,” in Proc. the 19th International ACM SIGACCESS Conference on Computers and Accessibility, Baltimore, USA, 2017. 4. P. Shneha, P. Reddy, and V. M. Megala, “Artificial intelligence for vision impaired people,” International Journal of Latest Trends in Engineering and Technology, pp. 031–036, 2018. 5. L. Deng, “Artificial intelligence in the rising wave of deep learning: the historical path and future outlook,” IEEE Signal Processing Magazine, vol. 35, no. 1, pp. 180–177, 2018. 11

Volume 1, Issue 2 June 2024 12 Authors Abdinabi Mukhamadiyev Department of Computer Engineering, Gachon University, Sujeong-gu, Seongnam-si 13120, South Korea; mukhamadiyev@ gachon.ac.kr Jinsoo Cho Department of Computer Engineering, Gachon University, Sujeong-gu, Seongnam-si 13120, South Korea; jscho@gachon.ac.kr *Correspondence: mukhamadiyev@ gachon.ac.kr Management and Future Technologies journal.umft.uz UZASSISTANT: REVOLUTIONIZING CUSTOMER SERVICE IN UZBEKISTAN WITH AI A bstract: This paper introduces UzAssistant, an innovative artificial intelligence system designed to automate customer service interactions in Uzbek. By utilizing state-of-the-art speech recognition and synthesis technologies, UzAssistant aims to significantly enhance service efficiency and accessibility, addressing the linguistic needs of the Uzbek-speaking population. Preliminary results demonstrate high accuracy in voice recognition and naturalistic speech synthesis, promising substantial improvements in customer experience and operational efficiency. Keywords: Speech technologies, call center, Uzbek language, speech-to-text, text-to-speech, speech recognition, speech synthesis. Copyright: © 2024 by the authors. Ushbu maqola Creative Commons Attribution (CC BY) litsenziyasi shartlari asosida tarqatiladigan ochiq foydalanish maqolasi hisoblanadi (https://creativecommons.org/ licenses/by/4.0/).

Management and Future Technologies journal.umft.uz June 2024 Volume 1, Issue 2 Introduction 13 ǡ ʹͶȀ Ǥ ǡ Ǥ ǡ ǡ ȏͳǦȐǤ Methodology Ǥ ǣ ǡǡ ǡ ǦǦ ȋȌ ȏȐǤ ǡ ȋ ʹȌǤʹǡ ʹ Ǥǡ ǡ ǡ ǡ Ǧ ȏͺȐǤ Figure 1. ǡ Ǥ ǡǡ Ǧ

Volume 1, Issue 2 14 June 2024 Management and Future Technologies journal.umft.uz ȋ ȌǤ Ǧ ǡ Ǥ ̵ȏͻȐǤ Results ͻǤͶΨǡ Ǥ Ǥ ǡ ʹ ȋȌ ȋȌǤ ͷǡͶͶ ǡ ʹ ͳ͵ǤͺΨͷǤʹʹΨǡ Ǥ Table 1. ʹǦ Ǧ ʹǦ ʹ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ CER WER CER ͳ͵Ǥͻ Ͷ͵Ǥʹ ͳͷǤͳ ͶͶǤ ͳ͵Ǥͺ ʹǤ ͳͷǤͷ ͵ͳǤʹ ͳͶǤͺ ͳʹǤ ͳͲǤ ͳ͵Ǥͳ ͳͲǤͻ Ǥ͵ Ǥʹ ǤͲ ͻǤ͵ ͺǤʹ ͺǤͳ Ǥͻ Ǥ ͳʹǤͲ ͳͳǤ͵ ͺǤͻ Ǥʹ 5.4 ͵ͲǤͳ ʹͶǤͺ ʹʹǤ ͵ͷǤͳ ʹͳǤ͵ ͳͻǤͶ ʹͲǤʹ ͳǤͳ ͳͷǤʹ ͳʹǤ͵ ͳͳǤͷ ͳͲǤͻ ͻǤͲ ͺǤͳ ͺǤ Ǥͷ ͶͲǤͷ ͳʹǤ ͵ͲǤ͵ ͻǤͻ ͵ʹǤ ʹͶǤͳ ʹ͵Ǥͳ ͳͲǤ͵ ͻǤʹ ͺǤͻ ͵Ǥ ͳͲǤʹ ʹͲǤ Ǥͳ ʹǤͶ ͳǤͷ 15.1 ͻǤ͵ ͷǤͻ 5.22 WER ͵ͳǤͻ ʹǤͺ ʹ͵Ǥͻ ͵ʹǤ ʹͷǤͶ ʹ͵Ǥͻ ʹͷǤͶ ʹͳǤͻ ͶͶǤʹ ʹͺǤ ʹǤʹ ʹͶǤͶ ʹʹǤ͵ ͵ͶǤ ʹͷǤͻ ʹͲǤ ͳǤͻ 13.8

Management and Future Technologies journal.umft.uz June 2024 Volume 1, Issue 2 15 ǡ ̵ Ǥ Discussion automating routine interactions. Future work will focus on expanding UzAssistant’s domain of Ǥ Conclusion Ǧ ǡ Ǧ Ǥ ǡ Ǥ

Volume 1, Issue 2 16 June 2024 Management and Future Technologies journal.umft.uz References 1. Guzman, A. L. (2018). Voices in and of the machine: Source orientation toward mobile virtual assistants. Computers in Human Behavior, 90, 343–350. 2. McCue, T. J. (2018). Okay Google: Voice search technology and the rise of voice commerce. Forbes Online, Retrieved From: https://www.forbes.com/sites/tjmccue/2018/08/ 28/okaygoogle-voice-search-technology-and-the-rise-of-voice-commerce/# 57eca9124e29 3. Juniper Research (2018). Voice Assistants used in smart homes to grow 1000%, reaching 275 million by 2023, as Alexa leads the way. Retrieved from: https://www. juniperresearch.com/ press/press-releases/voice-assistants-used-in-smart-homes. 4. Mukhamadiyev, A.; Khujayarov, I.; Djuraev, O.; Cho, J. Automatic Speech Recognition Method Based on Deep Learning Approaches for Uzbek Language. Sensors 2022, 22, 3683 5. Mukhamadiyev, A., Mukhiddinov, M., Khujayarov, I., Ochilov, M., & Cho, J. (2023). Development of Language Models for Continuous Uzbek Speech Recognition System. Sensors, 23(3), 1145. 6. Smith, S. (2018). Voice assistants used in smart homes to grow 1,000%, reaching 275 million by 2023, as alexa leads th. Juniper Research. Retrieved July 2021 from https://www.juniperresearch. com/press/voice-assis tants-in-smart-homes-reach-275m-2023. 7. U. Khamdamov, M. Mukhiddinov, B. Akmuradov and E. Zarmasov, “A Novel Algorithm of Numbers to Text Conversion for Uzbek Language TTS Synthesizer,” 2020 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, 2020, pp. 1-5, doi: 10.1109/ICISCT50599.2020.9351434. 8. Oord, A.V.D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A. and Kavukcuoglu, K., 2016. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499 9. E. Song, F. K. Soong, and H.-G. Kang, “Effective Spectral and Excitation Modeling Techniques for LSTM-RNN-Based Speech Synthesis Systems,” IEEE/ACM Trans. Audio, Speech, and Lang. Process., vol. 25, no. 11, pp. 2152–2161, 2017.

Management and Future Technologies journal.umft.uz ENHANCED DIAGNOSIS OF CATTLE DISEASES USING FUZZY LOGIC ALGORITHMS June 2024 Volume 1, Issue 2 17 Authors Dilmurod Turimov Department of IT Convergence Engineering, Gachon University, Sujeong-Gu, Seongnam-Si 461-701, Gyeonggi-Do, Republic of Korea dilmurod@gachon.ac.kr A bstract. Cattle illnesses can greatly affect both animal health and the economic stability of farmers. Prompt and precise diagnosis is crucial for effective management and treatment of these diseases. This study explores the development of diagnostic models and algorithms for cattle diseases using Sugeno‘s fuzzy inference system. It includes an analytical review of mathematical methods for animal disease diagnosis and soft computing approaches for tackling classification challenges. The study proposes an algorithm to create a knowledge base from the clinical signs of diseases, enhancing the reliability of informative features. Utilizing this algorithm, a software program was developed to diagnose cattle diseases. A computational experiment was conducted, and its results now serve as additional decisionmaking tools for diagnosing diseases in cattle. Using the program, a Sugeno fuzzy logic model was constructed and its results’ adequacy analyzed. The study also evaluates the performance of this model by comparing it with existing algorithms and addresses various classification and evaluation challenges. The outcomes facilitate quick diagnosis and timely intervention, reducing data analysis time and improving the diagnostic process’s efficiency. The scientific innovation of this research lies in the development of a knowledge base construction algorithm and enhancements to the Sugeno fuzzy logic model for cattle disease diagnosis. The findings have significant implications for veterinary medicine, particularly in the diagnosis of cattle diseases and supporting decision-making in intelligent systems. Keywords: osteodystrophy, secondary, steodystrophy, ketosis, hypomicroselementosis, decision making, expert systems, fuzzy sets. Copyright: © 2024 by the authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/ by/4.0/).

Volume 1, Issue 2 18 June 2024 Management and Future Technologies journal.umft.uz Introduction Ǥ ǡ ȋ Ȍ Ǥ Ǧ ǡ ǡ Ǥ ǡ Ǥ ǡ ǡ Ǥ ǡǡ Ǥ ǡǡ Ǥ ǡ ǡ Ǥ Ǧ ǡ ǡ Ǥ ǡǡ Ǥ ǡ ǡ ǡ ǡǡǤ ǡ ǡ ǡǤ ǡ ǡ Ǥ Methods Ǧbased system using Sugeno’s Ǥ ǡ ǡ Ǥ ǡ ǤͶͷͲ Ǥ

Management and Future Technologies journal.umft.uz June 2024 Volume 1, Issue 2 19 ʹǤ Figure 1. Figure 2. Ǥ

Volume 1, Issue 2 20 June 2024 Management and Future Technologies journal.umft.uz Results Ǥ ǡ Ǥ ǡ ͻͺΨǡ Ǥ Conclusion ǡ ǡǤ Ǧ Ǥ Ǥ