Intelligent Data Analysis as a Method of Determining the Influence of Various Factors on the Level of Customer Satisfaction of the Company

In this study, a neural network model was developed and investigated for predicting crop yields based on data on weather conditions, the use of fertilizers and the content of basic nutrients in the soil (nitrogen, phosphorus and potassium). The research is based on the use of a multilayer perceptron architecture with Rely activation functions for hidden layers and linear activation for the output layer. The evaluation of the model quality was carried out using the mean square error (MSE), which was 0.5783 in the test sample, demonstrating high accuracy of predictions. Visualization of the results included analysis of scatter plots, residuals, histograms of residuals and comparison of distributions of actual and predicted values. The results obtained confirm the effectiveness of the proposed model for yield forecasting tasks, which makes it a valuable tool for optimizing agricultural production.

Crop yield forecasting using neural networks trained on the basis of agrometeorological and agrochemical data

Degtyareva,

Tynchenko,

Stepanov

et al. 2024

Application of machine learning methods to predict soil moisture based on meteorological and atmospheric data

Tynchenko,

Kukartseva,

Degtyareva

et al. 2024

The purpose of this study was to develop and evaluate models for predicting soil moisture based on data from meteorological conditions and particle concentrations in the air. Two machine learning methods were used in the work: random forest and linear regression. The results of the study showed that the random forest model achieved 94% accuracy, while the linear regression model showed 92% accuracy. Air temperature, air humidity and the concentration of particles in the air turned out to be important factors affecting soil moisture. Both models offered good predictive capabilities, with an emphasis on the ability of a random forest to adapt to complex nonlinear dependencies, and linear regression to interpret the results. The developed models can be useful for optimizing agricultural processes, managing land resources and environmental monitoring.

Development of a model for predicting soil moisture dynamics

Degtyareva,

Tynchenko,

Kozlova

et al. 2024

This paper discusses the development of a model for predicting soil moisture dynamics based on remote sensing data and soil characteristics using neural networks. In the course of the study, preliminary data processing was carried out, including scaling of features and analysis of correlations between them. The constructed model showed high accuracy of predictions, which is confirmed by the values of the standard error (0.00849) and the coefficient of determination (0.854). The test results demonstrated the ability of the model to effectively reproduce the actual values of soil moisture, which makes it a useful tool for water management and planning of agrotechnical measures. In conclusion, possible ways to further improve the model and expand its application are discussed.