Development of an algorithm for selecting repeat crops type and predicting yield in the conditions of agriculture in Uzbekistan

The classification challenges in agricultural decision-making are tackled through the development of a rule-enrichment algorithm based on artificial intelligence and knowledge graph techniques. This study focused on the selection of an appropriate second season crop under Uzbekistan’s agricultural condition using knowledge base graphs. In order to solve this problem, agro-climatic conditions were systematically collected and represented through the construction of graph - based knowledge base (KB) structures. After collecting data, the dataset was structured and processed to enable the extraction of the predicates (logical rules). The study also demonstrates a data-normalisation workflow that supports rule extraction and knowledge graph construction for classifying second-season crop types. A key contribution of this work is the implementation of a numerical-predicate enrichment algorithm that enhances the KB and supports accurate rule formation. The proposed enrichment algorithm substantially improved both rule quality and classification performance, as demonstrated by the detailed results presented in this study. The algorithm was partially tested through a software prototype deployed on agricultural datasets from the Jizzakh region, demonstrating its practical applicability.

Hali tarjima qilinmagan