Жахондаги олиб борилаётган илмий тадқиқотлардан кўришимиз мумкинки сунъий интеллект амаллари турли хил усуллар ёрдамида амалга оширилади, буларнинг ичида машинавий ўқитиш энг кенг тарқалган усул ҳисобланади. Бугунги кунда машинавий ўқитишни назоратли ўқитиш (supervised learning), назоратсиз ўқитиш (unsepervised learning), кучайтирилган ўқитиш (reinforcyement learning) тоифалари мавжуд. Машинали ўқитишда регрессия усулларидан чизиқли регрессия, кўп ўзгарувчили чизиқли регрессия ва полиномиал регрессия усуллари кенг қўлланилади. Ушбу мақолада машинали ўқитишда қўлланиладиган полигармоник сплайн моделларидан фойдаланилган. Дастлаб полиномли ва полином бўлмаган сплайнларни солиштириш амалга оширилган. Полигармоник сплайнларни интерполяциялаш жараёнларига мисоллар келтирилган. Полигармоник спайн билан интерполяция қилишнинг асосий афзаллиги ва камчиликлари келтириб ўтилган.
Sign language recognition has gained significant attention due to its potential to bridge communication gaps between the deaf and hearing communities. This article presents a comprehensive review of machine learning methods employed for the recognition of Uzbek Sign Language (UzSL). The unique visual and spatial nature of sign languages poses challenges that necessitate specialized techniques for accurate recognition. This review surveys various approaches, ranging from traditional techniques to modern deep learning methods, used to recognize UzSL gestures. The article begins by introducing the significance of UzSL recognition and its impact on facilitating effective communication for the Uzbek deaf community. It outlines the complexities involved in sign language recognition, including variations in hand shapes, movements, and facial expressions. The challenges of limited training data, real-time recognition, and capturing dynamic features are discussed in depth. A survey of traditional machine learning methods such as Hidden Markov Models (HMMs), Support Vector Machines (SVMs), and k-Nearest Neighbors (k-NN) is presented, along with their applications and limitations in UzSL recognition. The evolution of these methods into more sophisticated approaches like Dynamic Time Warping (DTW) and Conditional Random Fields (CRFs) is also explored.
Ushbu maqolaning maqsadi mashinaviy o‘qitish (machine learning) usullaridan foydalangan holda maktab o‘quvchilarini fanlardan olgan baholari va softskills ko‘nikmalari bo‘yicha ularning yo‘nalishlarini aniqlashdir. Maqolada ko‘p o‘zgaruvchili chiziqli regressiya yordamida sigmoid funksiya qurish murakkabliklari ko‘rib chiqildi hamda o‘quvchilarni 10 yil davomida ta’lim yo’nalishi bo‘yicha tanlangan fanlardan olingan baholari va ularni turli parametrlari, sabablari hamda o‘quvchining imkoniyatlari raqamlashtirildi. Ushu raqamlar yordamida trening ma’lumotlar top‘lami tashkil qilindi. Natijada maktab o‘quvchilarini 10 yil davomida o‘qigan fanlari va ulardan oligan baholarining klassifikatsiyasi ishlab chiqildi. Neyron tarmoq arxitekturalari, modullari, mashinaviy o‘qitish algoritmlarida eng ko‘p qo‘llanilayotgan aktivlashtirish funksiyalari, o‘qitish usullari hamda chiziqli va logistik regressiya qurish usullari, kamchiliklari va imkoniyatlari tahlil qilingan. Ko‘p o‘zgaruvchili chiziqli regressiya uchun gradiyent tusish funksiyasini vektorlangan hisoblash orqali qulaylashtirish yo‘llari o‘rganib chiqildi. Chiziqli regressiyaning bu turida juda ko‘p o‘zgaruvchilar qatnashganligi uchun vektor hisob-kitoblar ancha qulayli isbotlangan. Vektor hisoblashlar yordamida gradiyent tushish jarayonlarini parallel hisoblash yo‘llari ham ko‘rib o‘tilgan. Hususan, trening ma’lumotlar jadvalining ustunlarini qo‘shish, koeffitsiyentlarni transpozitsiyalash - AT, chiziqli funksiyaning vektorlangan ko‘rinishlari, gradiyent tushish uchun giperparametrlar (o‘rganish darajasi - , qadamlar soni) aniqlab olindi.
Машинасозлик саноатнинг бошқа соҳалари ичида юқори технологиялардан фойдаланиш бўйича етакчилик қилади ва иқтисодиётда катта мультипликатив аҳамиятга эга. Иқтисодиётнинг бошқа соҳаларида илғор машиналар, қурилмалар ва технологик жараёнлар татбиқ этилишида муҳим роль айнан машинасозликка тегишли. Мақолада жаҳон иқтисодиётида, Шарқий Осиё мамлакатларида (Япония, ХХР) машинасозлик ривожланишининг замонавий тенденциялари таҳлил қилинган, янги индустриал мамлакатлар тажрибасида соҳанинг Ўзбекистонда тараққиёт истиқболлари кўриб чиқилган.
The remarkable development of accessible data sources has enormously impacted the admittance to useable wellbeing data. As an outcome, restoratively one-sided data has become hard to use for navigation. In this paper, we consider these outcomes and present an improved technique for getting to wellbeing data continuously. The methodology includes the utilization of the vapnik Backing Vector Machine process for text grouping. The proposed technique was frameworked on php/mysql for web client. Trial arrangement shows that the strategy outflanks the pattern in the Accuracy, Review and F1 measures. An expansion utilizing the Gaussian portion is suggested in the paper.
Machine learning algorithms play a crucial role in extracting valuable insights from data, enabling businesses and researchers to make informed decisions. One such algorithm is the decision tree, which is widely used for classification tasks. Decision tree classification utilizes a tree-like model of decisions and their potential consequences, making it an intuitive and powerful tool for solving complex problems. In this article, a model that determines which drug is suitable for a patient with a certain disease is created using the Decision tree algorithm. This problem is multi-class classification (multiclass classification) help score consolidation. Alternatively, how function, domain, and hyperparameters simplify decision tree models are explored.