استخدام خــــــــوارزمية XGBoost لتنــــــــبؤ بالتضخم وأسعار الغذاء العالمية: تـــــحليل زمــــني حسب المنتج السوق والدولة"

Abstract This study aims to utilize the XGboost algorithm to predict global inflation and food prices, based on data collected from the Kaggle database. The model was evaluated using a set of visual and performance metrics, which were comprehensively analyzed in this study. A diverse dataset was used to train and evaluate the model, with the goal of analyzing temporal data to identify patterns and relationships between different variables. The results demonstrated high accuracy for the model in predictions, with most points being close to the ideal red line. The model's performance was evaluated using the following metrics: (MSE): 12.8% (RMSE): 35.7% (MAE): 5.4% R² Score: 99.5% These results reflect the model's efficiency in estimating food prices and inflation, providing reliable and effective recommendations for decision-makers. The model's performance can be further enhanced by conducting additional studies to explore different architectures and improve its accuracy.