This paper introduces an AI-driven analytical framework designed to automatically identify manipulative linguistic strategies embedded in Telegram user comments. The platform’s open communication structure, minimal moderation, and multilingual user base create favorable conditions for the spread of manipulative discourse, often crafted to influence emotions, beliefs, and decision-making processes. Unlike traditional text classifiers, the proposed framework models manipulation as a complex phenomenon combining rhetorical, psychological, and contextual signals. The detection system integrates a compact transformer architecture with discourse-level features such as inferential intent, persuasive constructions, illusion-of-choice tactics, and emotionally charged framing. A specially created dataset of Telegram comments from regional channels was annotated using a multi-label scheme capturing diverse manipulation types. Experimental evaluation demonstrates that the proposed model achieves reliable performance in distinguishing manipulative comments from neutral communication despite noise, code-switching, and message brevity. The system’s design allows for scalable deployment in moderation tools, research platforms, and automated risk-analysis pipelines.
13.12.2025 Ko`rish Yuklab olish
Methods for intelligent analysis of soil composition data based on modern artificial intelligence algorithms are important in increasing agricultural efficiency and ensuring a sustainable ecology. The study examined methods for analyzing soil data using machine learning algorithms to overcome some of the shortcomings of traditional laboratory methods for this process. The most modern methods were considered, from processing datasets to training them in artificial intelligence models. These methods can be used to perform real-time analysis processes in agriculture.
13.12.2025 Ko`rish Yuklab olish
Bugungi kunda jahon miqyosida raqamli iqtisodiyot, innovatsiya va axborot texnologiyalari shiddat bilan rivojlanmoqda. Ushbu jarayonning asosiy drayverlaridan biri - sun’iy intellekt (SI) hisoblanadi. U insonning fikrlash, tahlil qilish, xulosa chiqarish va qaror qabul qilish qobiliyatini taqlid qiladigan tizimlarni yaratishni maqsad qiladi. Sun’iy intellekt nafaqat texnika, balki ijtimoiy hayot, fan, ta’lim va tibbiyot sohalarida ham keng qo’llanilmoqda.
13.12.2025 Ko`rish Yuklab olish
We present a hybrid artificial intelligence (AI) approach for early breast cancer detection that jointly leverages genetic, imaging, and clinical data. The model comprises three modality-specific encoders (genetic, mammography/MRI/ultrasound, and structured clinical variables) whose embeddings are fused at the feature level and fed to a unified classifier. In a proof-of-concept evaluation on a multi-modal cohort, the hybrid model achieved ~0.93 ROC-AUC with ~94% sensitivity and ~88% specificity, outperforming single-modality baselines. We discuss design choices (fusion strategy, robustness to missing modalities), limitations (data availability, explainability), and directions for clinical integration.
13.12.2025 Ko`rish Yuklab olish
This study analyzes the effectiveness of categorical data encoding methods in artificial intelligence algorithms. The research examines the operational characteristics and impact on results of four widely used encoding techniques—Label Encoding, One-Hot Encoding, Frequency Encoding, and Target Encoding—applied to Decision Tree and K-Nearest Neighbors (KNN) algorithms. Using a real-world dataset, each encoding method was applied separately and evaluated with both algorithms. Model performance was assessed using conventional evaluation metrics such as accuracy, precision, recall, and F1-score. The results indicate that the combination of encoding method and selected algorithm has a significant effect on model quality. In particular, One-Hot Encoding yielded the best results with Decision Trees, while Target Encoding was found to be most effective for the KNN algorithm. The study concludes by outlining important considerations and practical recommendations for selecting appropriate encoding methods.
13.12.2025 Ko`rish Yuklab olish
The following paper presents an integrated scientific analysis uniting three recent Scopus-indexed research directions: (1) AniFaceDiff - diffusion-based avatar animation, Pattern Recognition, 2026 (2) k-real graph complexity in visual structure modeling (Theoretical Computer Science, 2025), and (3) Public trust in AI-driven face recognition (Computers in Human Behavior, 2026). A comprehensive theoretical and experimental framework is proposed that combines Stable Diffusion-based geometric modeling with graph-theoretic optimization and ethical evaluation metrics derived from social data. The results indicate a 15–20% improvement in image fidelity and a 25% increase in perceived algorithmic fairness. The present framework unifies technical accuracy, mathematical transparency, and social legitimacy in AI-driven biometric systems.
13.12.2025 Ko`rish Yuklab olish
Мазкур мақола ўзбек тилидаги матнлардаги ҳиссиёт таҳлилини амалга оширувчи чуқур ўрганиш модели самарадорлиги таҳлил қилинган. Тадқиқотда матн таснифлаш вазифасида кўп вазифали ўқитиш ёндашуви асосида ишлаб чиқилган MTL-BERT модели таклиф этилди. Модел ижтимоий тармоқлар ва янгилик сайтларидаги фойдаланувчи изоҳларини сентимент таҳлил қилиш учун қўлланилди. Натижалар шуни кўрсатдики, таклиф этилган модель анъанавий ёндашувларга нисбатан юқори аниқликка эга бўлиб, F1 кўрсаткичи 93% гача етди. Бу ўзбек тили учун чуқур ўрганишга асосланган ҳиссиёт таҳлилининг самарадорлигини тасдиқлайди.
13.12.2025 Ko`rish Yuklab olish
Ushbu ishda tuzilmalanmagan hujjat tasvirlarida qo‘lyozmali imzo sohasini avtomatik aniqlash va ajratib olish masalasi ko‘rib chiqilgan. Taklif etilgan algoritm morfologik amallar, adaptiv filtratsiya va segmentatsiya bosqichlarini o‘z ichiga oladi. Tadqiqotda 2500 dan ortiq imzolangan hujjat tasvirlari ustida tajriba o‘tkazilib, algoritmning samaradorligi aniqlash mezonlari orqali baholandi. Natijalar algoritmning yuqori aniqlikka ega ekanini, shuningdek, bosma matn, shtamp va fon bilan aralashgan holatlarda ham imzo sohasini to‘g‘ri ajrata olishini ko‘rsatdi. Taklif etilgan algoritm avtomatlashtirilgan hujjat tasnifi, imzo verifikatsiyasi va raqamli arxiv tizimlarida qo‘llanishi mumkin.
13.12.2025 Ko`rish Yuklab olish
Тщательное соблюдение агротехнических требований к влажности почвы является залогом высокой эффективности растениеводства. Поскольку влажность почвы зависит от вида растений, от периода вегетации и от многочисленных погодных факторов, важное значение приобретает автоматизация контроля и регулирования влажности почвы в корневой зоне растений. В докладе рассмотрены вопросы применения технологии IoT для контроля и регулирования влажности почвы и создания на её базе интеллектуальной системы орошения, позволяющей фермерам автоматизировать процессы полива и сократить потребление воды.
13.12.2025 Ko`rish Yuklab olish
Ushbu maqolada sun’iy intellekt (SI) xizmatlari eksportini rivojlantirishda davlatning qo‘llab‑quvvatlash mexanizmlari va innovatsion autsorsing siyosatining ahamiyati tahlil qilinadi. O‘zbekiston Respublikasining amaldagi dastur va rag‘batlari o‘rganilib, Hindiston, Polsha va Estoniya kabi ilg‘or davlatlar tajribasi bilan solishtirish asosida takomillashtirish bo‘yicha amaliy tavsiyalar ishlab chiqilgan. Tadqiqotda statistik tahlil, komparativ metod, ekspert bahosi va siyosiy‑institutsional baholash usullari keng qo‘llanildi.
13.12.2025 Ko`rish Yuklab olish