OJS Tashkent State University of Economics
Not a member yet
554 research outputs found
Sort by
INNOVATSION RAQAMLI MUHITDA SAVDO KORXONALARINING AXBOROT TIZIMINI RIVOJLANTIRISH YO‘LLARI
No full textUshbu maqolada innovatsion raqamli muhitda savdo korxonalarining axborot tizimini rivojlantirish yo‘llari masalasi keng qamrovda o‘rganiladi. Tadqiqotda savdo korxonalarida raqamli texnologiyalarni joriy etish, ma’lumotlar bazasi va axborot oqimlarini samarali boshqarish, biznes jarayonlarni avtomatlashtirish hamda raqobatbardoshlikni oshirish imkoniyatlari tahlil qilinadi. Milliy va xorijiy tajribalar asosida savdo korxonalarida axborot tizimining iqtisodiy samaradorligi, uning boshqaruv qarorlarini qabul qilishdagi roli va innovatsion yechimlari ko‘rib chiqiladi. Natijalarga ko‘ra, kompleks model savdo korxonalarining faoliyatini raqamlashtirish, xarajatlarni kamaytirish va iste’molchilarga xizmat ko‘rsatish sifatini oshirishda muhim ahamiyat kasb etishi aniqlangan. Maqola yakunida O‘zbekistonda savdo korxonalari uchun axborot tizimini rivojlantirishning ustuvor yo‘nalishlari bo‘yicha takliflar ishlab chiqilgan
DEEP LEARNING APPROACHES IN CLASSIFICATION OF ECG SIGNALS FOR CARDIOVASCULAR DISEASE DETECTION
No full textCardiovascular diseases (CVDs) remain the leading cause of mortality worldwide, emphasizing the critical need for early detection and accurate diagnosis. Electrocardiography (ECG) provides a noninvasive and cost-effective means of assessing cardiac health; however, manual interpretation is time-consuming and prone to human error. Recent advances in deep learning have shown great promise in automating ECG signal analysis by leveraging powerful feature extraction and classification capabilities. This study investigates multiple deep learning architectures, including convolutional neural networks (CNNs), hybrid CNN–LSTM models, and Transformer-based approaches, applied to benchmark ECG datasets. A comprehensive workflow is developed, consisting of data acquisition, preprocessing, beat segmentation, deep learning-based feature extraction, classification, and performance evaluation. Comparative experiments demonstrate that while CNNs effectively capture morphological features, the CNN–LSTM hybrid yields enhanced performance by modeling temporal dependencies. The Transformer-based model achieves the highest accuracy, sensitivity, and specificity, highlighting its ability to capture long-range dependencies in ECG signals. The results confirm the potential of advanced deep learning frameworks in supporting reliable, automated CVD detection. Future directions include cross-dataset validation, multimodal integration, and optimization for deployment on wearable and embedded platforms
COMPARISON OF THE EXACT AND ASYMPTOTIC SOLUTIONS TO THE PROBLEM OF OSCILLATORY FLOW OF A VISCOUS FLUID IN A CYLINDRICAL PIPE
No full textThis article explores the problems of oscillatory flow of a viscous incompressible fluid in a cylindrical pipe with a given harmonic oscillation are solvedaverage velocity across the pipe cross-sectionThe transfer function and amplitude-phase frequency characteristics were determined. Using these functions, the influence of oscillation frequency on the ratio of shear stress to the cross-sectional average velocity was determined. By comparing the exact solution of the problem with the asymptotic solution, new asymptotic formulas are proposed that are of great importance in the study of the oscillatory flow of a viscoelastic fluid in a cylindrical pipe
TARMOQ MA’LUMOTLARIGA INTELLEKTUAL ISHLOV BERISH MODELLARI VA USULLARI
No full textUshbu ishda tarmoq ma’lumotlariga intellektual ishlov berishning zamonaviy modellari va usullari tahlil qilingan. Tarmoq ma’lumotlarining hajmi va murakkabligi ortib borayotgani sababli, ularni samarali tahlil qilish va qayta ishlash uchun sun’iy intellekt, mashinali o‘qitish hamda ma’lumotlar mayning (data mining) texnologiyalaridan foydalanish zarurligi asoslab berilgan.Tadqiqotda tarmoq ma’lumotlarining tuzilishi, ularning turlari (ijtimoiy tarmoqlar, kompyuter tarmoqlari, sensor tarmoqlari va boshqalar) hamda ma’lumot oqimlari bilan bog‘liq muammolar yoritilgan. Shuningdek, intellektual ishlov berishda qo‘llaniladigan asosiy usullar — klassifikatsiya, klasterlash, prognozlash, anomaliyalarni aniqlash va graf analizi usullarining afzalliklari va qo‘llanilish sohalari keltirilgan.Ishda shuningdek, neyron tarmoqlar, graf neyron tarmoqlari (GNN), konvolyutsion va rekurrent modellardan foydalanish imkoniyatlari hamda ularning tarmoq ma’lumotlari tahlilidagi samaradorligi tahlil qilingan. Natijada, intellektual ishlov berish usullari orqali tarmoq ma’lumotlaridan qimmatli ma’lumotlarni avtomatik tarzda ajratib olish, xavfsizlikni ta’minlash va boshqaruv qarorlarini qabul qilishni takomillashtirish mumkinligi ko‘rsatilgan
IMPROVING ENERGY EFFICIENCY OF HVAC SYSTEMS USING ARTIFICIAL INTELLIGENCE METHODS
No full textThis paper addresses the improvement of energy efficiency in Heating, Ventilation, and Air Conditioning (HVAC) systems within large commercial buildings. It discusses the development and implementation of an Artificial Intelligence AI control platform integrated into the existing Building Management System BMS. A hybrid control algorithm is proposed, combining Model Predictive Control MPC, Reinforcement Learning RL, and Trim & Respond heuristics. This approach solves a multi-criteria optimization problem: minimizing energy consumption while strictly adhering to thermal comfort and Indoor Air Quality IAQ standards. The effectiveness of the approach is demonstrated through a pilot implementation in a large shopping and entertainment center. The results indicate a 23% reduction in ventilation electricity consumption compared to the baseline scenario, while maintaining CO₂ levels and temperature within regulatory ranges. The findings are benchmarked against international studies, confirming the viability of hybrid AI strategies without the need for major equipment retrofits
MODAL GRAPH TRANSFORMER FOR GROUP ENGAGEMENT IN ONLINE MULTIPARTY CONVERSATIONS
No full textOnline multiparty conversations such as virtual classrooms, remote collaboration meetings, and live discussions pose unique challenges for understanding group engagement. Traditional models typically rely on isolated participant features or unimodal data, failing to capture the rich, relational dynamics across modalities. In this work, we propose a novel approach that models group engagement as a dynamic, multimodal graph learning problem. Our framework introduces a Multimodal Graph Transformer (MGT) that combines audio-visual fusion with structure-aware graph attention. Each participant is represented as a graph node enriched with fused video and speech embeddings, while edges capture interaction intensity through gaze, turn-taking, and vocal overlap. To preserve graph structure, we incorporate hop-level positional encodings and restrict attention to top-k neighbors for scalable relational modeling. The architecture is designed to capture both localized interaction cues and global group dynamics without requiring pre-defined templates or scripted behavior. By integrating techniques from graph representation learning, multimodal attention, and social signal processing, our method offers a generalizable and theoretically grounded framework for engagement estimation in complex multiparty scenarios
KUTUBXONA TIZIMLARIDA FOYDALANUVCHILARGA AXBOROT EHTIYOJLARIGA KO‘RA RESURSLARNI TAVSIYA ETISH
No full textHozirda kutubxona tizimlari jamiyatning axborotga bo‘lgan ehtiyojlarini qondirishda muhim o‘rin tutadi hamda ular katta foydalanuvchilar auditoriyasiga ega. Ushbu tizimlar katta hajmdagi qimmatli axborot resurslarini o‘z ichiga olgan bo‘lib, foydalanuvchilarga zarur adabiyotlarni topishda yordam beruvchi qidiruv xizmatlarini ham taklif etadi. Qidiruv tizimlari foydalanuvchi so‘rovlarini qayta ishlaydi va mos deb topilgan natijalarni taqdim etadi. Biroq, bu jarayon natijasida foydalanuvchilar ko‘plab ma’lumotlar orasidan o‘zlariga eng mosini tanlashda qiyinchiliklarga duch kelishadi. Shu bois, zamonaviy axborot tizimlarida sun’iy intellektga asoslangan tavsiya etish tizimlari keng qo‘llanilib, foydalanuvchiga mos obyektlarni samarali tarzda taklif etish imkonini bermoqda. Kutubxona tizimlariga bunday tavsiya etish tizimi vositalarini joriy etish orqali foydalanuvchilarning axborot ehtiyojlarini aniqlash va ularga axborot ehtiyojiga ko‘ra resurslarni taqdim etish mumkin. Ushbu maqolada tavsiya etish tizimlari asosida foydalanuvchilarning axborot ehtiyojlarini aniqlash va resurslarni taqdim etishning nazariy asoslari hamda amaliy yondashuvlari tahlil qilinadi
AXBOROT TEXNOLOGIYALARI ASOSIDA DAVLAT XARIDLARIDA SHAFFOFLIK VA SAMARADORLIKNI TA’MINLASH
No full textMazkur maqolada budjet buyurtmachilari tomonidan tovar, ish, xizmatlar sotib olishda davlat xaridlarining zaruriyati ko‘rsatilgan. Davlat xaridlari ko‘lamining kengayishi mahalliy ishlab chiqaruvchilarni qo‘llab-quvvatlash, ish o‘rinlari yaratish, soliq tushumlarini oshirish bilan bog‘liqligi tahlil qilingan. Raqamli texnologiyalardan foydalanish asosida davlat xaridlarining ochiqligi va samaradorligini ta’minlash bo‘yicha amaliy taklif va tavsiyalar ishlab chiqilgan
Z-SONLАRDA NOАNIQLIK VА ISHONCHLILIKNI BIRGАLIKDА IFODАLАSHNING ZАMONАVIY YONDАSHUVI
No full textUshbu maqolada Lotfi Zade tomonidan taklif etilgan Z-sonlar konsepsiyasi haqida so‘z boradi. Z-sonlar ikki qismdan - qiymatning noravshan tavsifi va unga bo‘lgan ishonch darajasidan - iborat bo‘lib, kundalik hayotda uchraydigan taxminiy va subyektiv ma’lumotlarni hisobga olish imkonini beradi. Maqolada Z-sonlarning matematik tavsifi, turlari va ularni tavsiflash usullari yoritiladi. Shuningdek, ushbu yondashuvning iqtisodiyot, xavf-xatar tahlili va boshqaruv sohalaridagi amaliy qo‘llanish imkoniyatlari, afzalliklari hamda mavjud muammolari tahlil qilinadi
INFLATSIYANI BOSHQARISH USULLARINI TAKOMILLASHTIRISH VA INVESTITSIYAVIY MUHITNI RIVOJLANTIRISH YO‘NALISHLARI
No full textMaqolada inflyatsiyani boshqarish usullarini takomillashtirish va investitsiyaviy muhitni rivojlantirishning nazariy va amaliy jihatlari keng yoritilgan. Avvalo, inflyatsiya jarayonlarining shakllanishiga ta’sir etuvchi asosiy omillar, xususan talab va taklif inflyatsiyasi hamda inflyatsion kutilmalar mexanizmlarining iqtisodiyotni boshqarishdagi o‘rni tahlil qilindi. Tadqiqotda pul-kredit va fiskal siyosatning inflyatsiyani jilovlashdagi roli, monetar instrumentlarni optimallashtirish, byudjet intizomini mustahkamlash, soliq tizimini soddalashtirish kabi yo‘nalishlar asosiy e’tiborda bo‘ldi. Shuningdek, narx barqarorligini ta’minlashda raqobat muhitini rivojlantirish, bozordagi monopollashuv darajasini pasaytirish va iqtisodiyotning taklif qismini mustahkamlashning ahamiyati ko‘rsatib o‘tildi