OJS Tashkent State University of Economics
Not a member yet
554 research outputs found
Sort by
TA’LIM JARAYONIDA RAQAMLI TEXNOLOGIYALAR VA SUN’IY INTELLEKTNI QO‘LLASHNING ILMIY-ASOSIY FARQLARI VA TAHLILI
No full textMazkur maqolada zamonaviy ta’lim tizimida raqamli texnologiyalar va sun’iy intellekt (SI) texnologiyalarining o‘rni, ularning o‘zaro farqlari va ilmiy asoslari tahlil qilinadi. Raqamli texnologiyalar ta’lim infratuzilmasini raqamlashtirishga xizmat qilsa, sun’iy intellekt shaxsiylashtirilgan, moslashuvchan va tahliliy o‘quv muhiti yaratadi. Maqolada xalqaro va mahalliy tadqiqotlar asosida bu ikki texnologiyaning tarkibiy qismlari, qo‘llanilish sohalari, metodologik yondashuvlari va ta’limga ta’siri qiyosiy tahlil qilindi. Shuningdek, OECD, UNESCO, McKinsey kabi nufuzli tashkilotlarning statistik ma’lumotlari asosida ta’limda texnologik transformatsiya jarayonlarining real holati va istiqbollari baholandi
TADBIR TURIZMI ORQALI MAVSUMIY TURIZM OQIMLARINI MUVOZANATLASHTIRISH VA IJTIMOIY FAROVONLIKNI OSHIRISH IMKONIYATLARI
No full textTadbir turizmi turli yo‘nalishlarda turli xil samaradorlikka ega bo‘lsa-da, turizm mavsumiyligini yumshatish uchun strategik yondashuv sifatida paydo bo‘ladi. Tadqiqotlar shuni ko‘rsatadiki, tadbirlar yil davomida turizmga bo‘lgan talab taqsimotini muvozanatlashda yordam berishi mumkin, buni Makedoniyaning Sharqiy mintaqasida ko‘rish mumkin, bu yerda tadbir turizmi mavsumiylikning past darajasi va muvozanatli tashrif buyuruvchilar oqimiga hissa qo‘shgan [1]. Biroq, tadbirlar va mavsumiylikni boshqarish o‘rtasidagi bog‘liqlik murakkab bo‘lib, yil bo‘yi attraksionlarning atigi 39 foizi maxsus tadbirlarga mezbonlik qiladi, ayniqsa mahalliy hamjamiyatlarga mo‘ljallanmagan davrlarda [2]. Mavsumga qarshi turizm strategiyalari, shu jumladan madaniy, diniy va konferensiya turizmi bilan bir qatorda tadbirlar turizmi sayyohlarning kelishi va xarajatlaridagi vaqtinchalik nomutanosibliklarni bartaraf etish uchun tizimli yondashuvlarni ifodalaydi [3]. Shunga qaramay, kuchliroq mavsumiylik namunalari ishonchsiz turizm talabi prognozlari bilan bog‘liq bo‘lib, qo‘llaniladigan yumshatish strategiyalaridan qat’i nazar, rejalashtirish harakatlarini murakkablashtiradi [4]. Ushbu topilmalar shuni ko‘rsatadiki, tadbir turizmi mavsumiylikni kamaytirishga hissa qo‘shishi mumkin, ammo uning samaradorligi strategik harakatlar va mahalliy bozor sharoitlariga bog‘liq
HYBRID AI FOR GASTRIC DIAGNOSTICS: MULTI-MODAL FUSION AND INTERPRETABILITY FOR EARLY ULCER AND CANCER DETECTION
No full textThis study investigates the development of hybrid Artificial Intelligence (AI) models for the early and accurate diagnosis of gastric ulcer and gastric cancer. Traditional diagnostic methods such as endoscopy and biopsy, though widely used, face limitations including inter-observer variability, sampling errors, and delays in diagnosis. To address these challenges, two novel hybrid AI architectures are proposed: the Deep Feature-Attention SVM (DFA-SVM) and the Multi-Modal Fuzzy-Fusion Network (M2F2-Net). These models integrate multi-modal data sources, including endoscopic images, histopathology slides, and structured clinical records, while employing attention mechanisms, fuzzy logic, and interpretable AI approaches to enhance diagnostic performance. Experimental evaluation using the conceptual GIMIC-12K dataset demonstrates that M2F2-Net achieves superior performance, particularly in early gastric cancer detection, by significantly reducing false negatives and improving sensitivity. The findings highlight the transformative potential of hybrid AI in improving patient outcomes, optimizing healthcare efficiency, and supporting explainable decision-making in clinical practice
OʻZBEK TILIDAGI MATNLARNI TAHLIL QILISHDA NOMLI OBYEKTNI TANIB OLISH (NER) YONDASHUVI
No full textUshbu maqolada o‘zbek tilidagi matnlarni tahlil qilish jarayonida nomlangan obyektlarni tanib olish (Named Entity Recognition-NER) yondashuvlari har tomonlama tahlil qilingan. Tadqiqot doirasida NER tizimlarining nazariy asoslari, qoida asoslangan, mashinali o‘qitishga asoslangan hamda gibrid yondashuvlar batafsil ko‘rib chiqilgan va ularning afzalliklari hamda cheklovlari solishtirma tahlil asosida yoritilgan. Shuningdek, turli tillar va sohalarda, jumladan tibbiyot, oziq-ovqat va axborot tizimlarida qo‘llanilgan zamonaviy NER modellarining ishlash samaradorligi tahlil qilingan. Tadqiqot natijalari NER tizimlarining katta hajmdagi matnli ma’lumotlardan semantik axborotni avtomatik ajratib olishdagi ahamiyatini ko‘rsatadi hamda kam resursli tillar, xususan o‘zbek tili uchun moslashuvchan va samarali NER modellarini ishlab chiqish istiqbollarini belgilab beradi
TASVIR SIFATINI BAHOLASH KO‘RSATKICHLARI
No full textTasvirlash sohasidagi yutuqlar tez sur’atlar bilan davom etmoqda va bozorga yangi hamda ilg‘or mahsulotlar doimiy ravishda kirib kelmoqda. Yangi texnologiyalar joriy texnologiyaga nisbatan yuqori sifatli tasvirlarni ishlab chiqarishini tekshirish uchun sifatni baholashning samarali bir turi talab etiladi. Bu o‘z navbatida, tasvir sifatini baholash ko‘rsatkichlarini keng qamrovli tahlil qilish hamda ularni har biri bo‘yicha ma’lumotlarni bitta joyga yig‘ishni dolzarb masalaga aylantiradi. Shuning uchun, mazkur ishda tasvir sifatini baholash ko‘rsatkichlarini tarixi va tavsifi, ularni tasniflanishi haqida ma’lumotlar keltirilgan. Shuningdek, soha bo‘yicha mavjud adabiyotlar tahlili, tasvir sifatini baholash ko‘rsatkichlarini yutuq hamda kamchiliklari keltirilgan. Mazkur tadqiqot ishini maqsadi sifatni baholash ko‘rsatkichlarini alohida guruhlarga ajratish hamda ularni subyektiv idrok bilan mos kelishini baholashdan iborat. Bunday tasniflash ma’lum bir muammo yoki buzilish uchun eng mos keladigan sifatni baholash ko‘rsatkichini tanlashda foydali bo‘lishi mumkin
Z-SONLAR NAZARIYASI VA ULARNI KLASSIK NOANIQ SONLARGA AYLANTIRISH ALGORITMI
No full textUshbu maqolada Z-son tushunchasi, uning matematik ta’rifi va grafik ifodasi bayon etiladi. Z-son — bu (A, B) ko‘rinishidagi tartiblangan juftlik bo‘lib, A komponenti noaniq o‘zgaruvchining haqiqiy qiymatlariga qo‘yilgan cheklovlarni, B komponenti esa ushbu qiymatlarning ishonchlilik o‘lchovini ifodalaydi. Tadqiqotda Z-sonlarning diskret va uzluksiz ko‘rinishlari, ularning ehtimollik taqsimotlari bilan aloqasi, shuningdek, +Z-son tushunchasi tahlil qilinadi. Asosiy e’tibor Z-sonlarni klassik noaniq sonlarga aylantirish algoritmiga qaratiladi. Taklif etilgan usulda B komponenti integrallash orqali aniq son shakliga o‘tkazilib, A komponenti bilan vaznli tarzda birlashtiriladi. Natijada, trapetsiya yoki uchburchak ko‘rinishdagi noaniq sonlar klassik noaniq sonlarga konvertatsiya qilinadi. Ushbu yondashuv noaniq baholashlarni optimallashtirish va qaror qabul qilish tizimlarida samarali qo‘llanishi mumkin
SELECTING THE OPTIMAL COMBINATION OF MEMBERSHIP FUNCTIONS FOR QUERIES BASED ON FUZZY LOGIC BY EVALUATING THE SENSITIVITY
No full textEfficient request distribution is one of the important factors for the stable and uninterrupted operation of modern information systems. In the context of increasing cloud computing, real-time services and multi-user environments, traditional distribution algorithms (e.g. Round Robin, Least Connection, Weighted Response Time) are losing their effectiveness due to insufficient consideration of uncertainty and dynamic changes in server state. In this context, there is a growing need for decision-making systems based on adaptive and state-sensitive approaches.Fuzzy logic-based models have a significant advantage in representing uncertainties in real-world environments, especially due to the ability to estimate system parameters through linguistic variables. However, the effectiveness of these systems directly depends on the shape and parameters of the membership functions, i.e. their semantic relevance and sensitivity. Using membership functions of the same shape for each linguistic value (for example, only a triangle) can reduce the accuracy of the system\u27s decision-making
THEORETICAL FOUNDATIONS AND EFFICIENT ALGORITHMS FOR FEATURE DETECTION AND EXTRACTION
No full textFeature detection and extraction play a critical role in numerous applications within artificial intelligence, computer vision, and data science. As datasets grow in complexity and volume, traditional feature extraction techniques often struggle to maintain efficiency and accuracy. This paper explores the theoretical foundations and efficient algorithms for feature detection and extraction, emphasizing the intersection of classical statistical methods and modern machine learning-based approaches. The study provides an in-depth analysis of key methodologies, including edge detection algorithms, deep learning-based feature extraction, and dimensionality reduction techniques such as Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE). Furthermore, the research highlights the challenges associated with computational efficiency, accuracy trade-offs, and real-time processing constraints. A critical gap in existing literature is identified—the disconnect between theoretical models and their practical implementations—which often hinders the optimization of feature extraction processes. By synthesizing current advancements and addressing these challenges, this paper aims to provide a comprehensive framework for improving feature detection and extraction methodologies, ultimately contributing to the development of more efficient and adaptable systems across various domains
KOMPYUTER TARMOQLARIDA MAXFIY MA’LUMOTLARNI AI ASOSIDAGI ANIQLASH METODLARI
No full textMazkur maqolada, kompyuter tarmoqlarida uzatilayotgan va saqlanayotgan konfidentsial axborotlarni AI (Sun’iy intellekt) metodlari yordamida aniqlash yondashuvlari to‘g‘risida fikr yuritiladi.Hozirgi murakkab va dinamik tarmoq muhitida an’anviy yondashuvlar asosida tarmoqdagi ma’lumotlarni aniqlash yoki sizib chiqishini oldini olish yetarlicha foyda bermayotgani sababli mashinali o‘rganish va chuqur o‘rganish modellari tahlil qilinadi. Tadqiqotda klassifikatsiya va anomal holatlarni aniqlash usullarining ustunligi shuningdek ularni real tarmoq muhitida qo‘llashga doir fikr yuritiladi. Natijalar AI asosidagi metodlar kompyuter tarmoqlarida axborot xavfsizligini oshirishda muhim ahamiyatga ega ekanligini ko‘rsatadi
TURIZMDA TRANSPORT XIZMATLARI SIFATINI BAHOLASH KO‘RSATKICHLARINING AHAMIYATI
No full textChekka hududlarning rivojlanish imkoniyatlariga, xususan, turizmni rivojlantirishga ta’sir qiluvchi asosiy muammolardan biri mobillik va transportning qulayligi uchun qoniqarli darajada tashkil etilmaganligidir. Shu sababli, ushbu hududlarning raqobatbardoshligini oshiradigan, shuningdek, transportning atrof-muhitga salbiy ta’sirini kamaytiradigan innovatsion va barqaror transport yechimlariga talab oshib bormoqda. Ushbu maqola turizm sohasida kutilayotgan transport xizmatlarining sifati, transport xizmatlari sifatini baholash ko‘rsatkichlari tizimi, turizmda transport xizmatining tasniflanishi kabilar yoritib berilgan