This paper presents a single-layer feedforward neural network for morphological stemming of Uzbek words. Each word is encoded using character-level one-hot vectors (810-dimensional input: 27 characters × 30 positions). The model uses sigmoid activation, MSE loss, and gradient descent backpropagation. A systematic hyperparameter search over epochs E ∈ {20,30,40}, learning rates η ∈ {0.5,0.6}, and thresholds θ ∈ {0.70–0.85} across 24 configurations on 3,441 word–stem pairs yields a peak accuracy of 98.34% (E=40, η=0.6, θ=0.70). The results demonstrate that minimal neural architectures can effectively model agglutinative morphology in under-resourced Turkic languages.
| Mualliflar | U. I. Salaev |
|---|---|
| Jurnal | Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi |
| Nashriyot | Самарқанд давлат университети |
| ISSN | 2091-5446, 2181-1296 |
| Nashr sanasi | 2026-07-01 |
| Jild | 2 |
| Son | 3 |
| Betlar | 135-141 |
| Til | Ingliz |
| DOI | 10.59251/2181-3973.2025.v1.138.1.4018 |
| Indeksga qo‘shilgan | 2026-09-02 |
DOI: 10.59251/2181-3973.2025.v1.138.1.4018 · Maqolaning asl sahifasi
| APA | U. I. Salaev (2026). A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS. Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi, 2(3), 135-141. https://doi.org/10.59251/2181-3973.2025.v1.138.1.4018 |
|---|---|
| GOST (OAK) | U. I. Salaev A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS // Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi. – 2026. – T. 2, № 3. – P. 135-141. – DOI: 10.59251/2181-3973.2025.v1.138.1.4018. |
| MLA | U. I. Salaev. “A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS.” Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi, 2(3), 135-141, 2026. https://doi.org/10.59251/2181-3973.2025.v1.138.1.4018 |
| Chicago | U. I. Salaev. “A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS.” Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi, 2(3), 135-141 (2026). https://doi.org/10.59251/2181-3973.2025.v1.138.1.4018 |
| Harvard | U. I. Salaev (2026) ‘A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS’, Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi, 2(3), 135-141. https://doi.org/10.59251/2181-3973.2025.v1.138.1.4018 |
| BibTeX | @article{uisalaev2026,
title = {A SINGLE-LAYER FEEDFORWARD NEURAL NETWORK MODEL FOR MORPHOLOGICAL STEMMING OF UZBEK WORDS},
author = {U. I. Salaev},
journal = {Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi},
year = {2026},
volume = {2},
number = {3},
pages = {135-141},
doi = {10.59251/2181-3973.2025.v1.138.1.4018},
url = {https://doi.org/10.59251/2181-3973.2025.v1.138.1.4018}
} |
Boshqa maqola uchun iqtibos tayyorlash
The paper describes the process of developing the use cases for the functional model of the “Scientific Expertise and Control” module within the digitalisation of scientific attestation processes. The module is intended…
Large language models (LLMs) are now embedded in many digital systems, which has exposed a new and distinctive attack surface. The most pressing threat is the prompt injection attack, in which adversarial text causes a…
This article investigates the magnetic and electronic properties of the magnesioferrite (MgFe₂O₄) mineral found in the rock formations of Uzbekistan using both experimental and computational methods. The magnetic…
In this paper, a problem of anomalous fluid filtration and solute transport is considered. Fractional derivatives are used to represent anomaly of the filtration process. Using finite difference method, the problem is…
Samarqand davlat universiteti ilmiy tadqiqotlar axborotnomasi — barcha maqolalar · 2026-yil arxivi