The article investigates the problem of determining semantic similarity of Uzbek texts using mathematical modeling. The study aims to integrate semantic analysis, vector space representation, similarity calculation, and decision-making based on a threshold value into a unified model. The methodology includes text preprocessing, semantic vector generation, and similarity measurement using cosine similarity. Experiments were conducted in the Python environment, and the results were evaluated using Accuracy, Precision, Recall, and F1 metrics. The findings show that vector-based approaches represent semantic relationships more effectively than lexical-statistical models, while threshold optimization reduces misclassification and improves the overall accuracy of text similarity detection systems.
| Mualliflar | Allaberganova, Nasiba |
|---|---|
| Jurnal | Al-Farg'oniy avlodlari |
| Nashr sanasi | 2026-06-07 |
| Son | 2 |
| Betlar | 282-286 |
| Til | Rus |
Semantik analiz, matn o‘xshashligi, matematik model, NLP, embedding, vektor fazo modeli, cosine similarity, semantik masofa, Chegaraviy qiymat modeli., Semantic analysis, text similarity, mathematical model, NLP, embedding, vector space model, cosine similarity, semantic distance, Boundary value model.
Ushbu maqolada veterinariya nazorati punktlarida biologik va epizootik xavfsizlikni ta'minlashning zamonaviy usullari "VIS-Border" axborot tizimidan amaliy tadqiqot sifatida foydalangan holda ko'rib chiqiladi. Mavjud…
This article analyzes the problem of pixel-level separation of crop plants from weeds in agricultural images. In this study, the semantic segmentation models DeepLabV3+, U-Net, and U-Net++ were evaluated under identical…
The article investigates the entry of water from a supply pipeline into a spiral chamber through water-discharge structures, examining the transformation of kinetic energy resulting from the emergence of angular…
This article proposes a quantum-resilient adaptive network security model for IoT and edge computing infrastructures. The model integrates firewalls, IDPS, Zero Trust, post-quantum cryptography, and resource-aware…
The article describes a PostgreSQL-based relational model for organizing automotive data on manufacturers, models, generations and trims. The analysis covers tables, enum types, foreign keys, temporal CHECK constraints…
The principle of development of a perfect multisensor system integrated with unmanned aerial vehicles for the purpose of accurate and operational monitoring of various objects is considered. The system collects current…
This article discusses in detail the theoretical foundations, structure and practical applications of fractal robot technology. The fractal concept is based on the principle of self-similarity, and this feature is…
Ushbu maqolada bir o'lchovli termoelastik jismning matematik modeli va uning sonli simulyatsiyasi muhokama qilinadi. Termoelastik jarayonlarni tavsiflovchi differentsial tenglamalar chekli farq usuli yordamida…
This article provides an in-depth analysis of advanced algorithms for automatic object detection and classification in video imagery. The rapid advancements in artificial intelligence and deep learning have…
In this research, machine learning algorithms including Logistic Regression, Random Forest, XGBoost, and Artificial Neural Networks were evaluated and compared. Experimental results demonstrated that the integrated…