Central AsianResearch Index

MACHINE LEARNING MODELS FOR YIELD FORECASTING AND WEED CLASSIFICATION IN CROP PRODUCTION: A CASE STUDY FROM THE KYRGYZ REPUBLIC

The article examines current trends in the development of artificial intelligence (AI) systems and their role in the transformation of various sectors of the economy, including agriculture. The aim of the work is to analyze the theoretical foundations of machine learning and neural networks, as well as the practical implementation of these methods to optimize processes in crop production. In the theoretical part, a comparative analysis of AI systems, machine learning methods, and neural network architectures (convolutional, recurrent) is conducted. Special attention is paid to the role of big data and cloud computing as drivers of the development of AI technologies. The practical significance of the research lies in the development and software implementation of machine learning models in Python. A multiple linear regression model was used to predict the yield of agricultural crops (potatoes, barley) in the conditions of farms in the Issyk-Kul region. The effectiveness of three variants of the gradient descent method is compared: batch, stochastic and mini batch. The problem of classifying weeds using pattern recognition technologies has also been solved, which makes it possible to differentiate weeds into juvenile, perennial and parasitic. The results of computational experiments confirm the applicability of the developed models to improve the efficiency of decision-making in the agricultural sector.

Сведения

Опубликовано
2026 · Т. 10-770, № 1/3, С. 126-133
Язык
Английский
ISSN
1694-7851 · 1694-8505
Издатель
Kyrgyz State University named after I. Arabaev
Аффилиации
Kyrgyz State University named after I. Arabaev