In this paper, we propose and evaluate a mathematical model that describes the reported data on crop pests to get an accurate prediction of production costs, food safety, and the protection of the environment. Meteorological factors are not the only things that affect a bumper harvest; it is also affected by crop plant diseases and insect pests. Studies show that relying solely on the naked-eye observations of experts to forecast well-planned agriculture is not always sufficient to achieve effective control. Providing fast, automatic, cheap, and accurate artificial intelligence-based solutions for that task can be of great realistic significance. The proposed approach is genetic programming (GP)-based and is explicitly directed at solving the symbolic regression of crop pest forecasting. The GP approach is used to create a fitted crop pest model. Our experimental results indicate that the GP model can significantly support an accurate and automatic building of a reliable mathematical model. Furthermore, a comparison between the GP model and a linear regression model is also provided. The developed GP model can successfully achieve a precision of approximately 0.0557.

In this paper, we propose and evaluate a mathematical model that describes the reported data on crop pests to get an accurate prediction of production costs, food safety, and the protection of the environment. Meteorological factors are not the only things that affect a bumper harvest; it is also affected by crop plant diseases and insect pests. Studies show that relying solely on the naked-eye observations of experts to forecast well-planned agriculture is not always sufficient to achieve effective control. Providing fast, automatic, cheap, and accurate artificial intelligence-based solutions for that task can be of great realistic significance. The proposed approach is genetic programming (GP)-based and is explicitly directed at solving the symbolic regression of crop pest forecasting. The GP approach is used to create a fitted crop pest model. Our experimental results indicate that the GP model can significantly support an accurate and automatic building of a reliable mathematical model. Furthermore, a comparison between the GP model and a linear regression model is also provided. The developed GP model can successfully achieve a precision of approximately 0.0557.