| Abstrakt: |
This study conducts an Exploratory Data Analysis (EDA) to analyze the effect of phosphorus on the yield of spring wheat, with the aim of developing a neural network model for prediction. The dataset used in this analysis consists of soil surface temperature, precipitation, humidity for April to September, and annual phosphorus application data from 2000 to 2022, obtained from the National Bureau of Statistics of the Republic of Kazakhstan. The literature review highlights previous studies on the effect of various factors on spring wheat yield, emphasizing the limited research specifically focusing on phosphorus in this context. The objective of this research is to perform EDA on the dataset to train a neural network model capable of accurately predicting the influence of phosphorus on spring wheat yield. The EDA process involves calculation of descriptive statistics and visualization techniques such as histograms, scatter plots, and correlation analysis. The results indicate a positive correlation between phosphorus application and spring wheat yield, with a few outliers. Additionally, soil surface temperature and precipitation show weak negative correlations, while humidity exhibits a weak positive correlation. These findings can assist in developing neural network model for predicting phosphorus effect on the yield of spring wheat. Further steps suggested for future research include dataset standardization or normalization, as well as a comparison of neural network training results with different data cleaning and preprocessing techniques. By addressing these aspects, this study aims to contribute to the advancement of crop yield prediction models and agricultural practices specific to spring wheat cultivation in Kazakhstan. [ABSTRACT FROM AUTHOR] |
