The Algorithm for Intelligent Analysis of Soil Composition Data Based on Filling NaN Values Using Matrix Factorization Method for Artificial Intelligence Models

This research is dedicated to solving the problems of determining its productivity and classifying suitable crops based on the intellectual analysis of soil composition data. The main focus of the study is on comparing methods for filling missing (NaN) values at the stage of dataset preparation. Traditional and modern imputation methods were compared based on the soil composition dataset. As a result, methods based on matrix factorization showed higher accuracy in classification algorithms such as KNN, SVM, random forest, decision tree, XGBoost compared to standard methods. With this approach, the working process of artificial intelligence models is optimized and the accuracy results are high. This research increases the need for artificial intelligence technologies in the agricultural sector to optimize the most important resources and prevent soil degradation.

Hali tarjima qilinmagan