Prediction of Food Wastage using XG Boost

This research study introduces a predictive modelling approach by employing machine learning algorithms to forecast food wastage levels, addressing significant challenges such as data variability, environmental dependencies, and socio-economic disparities. Recent techniques in food wastage prediction often struggle with integrating diverse datasets and achieving high predictive accuracy. Utilizing a comprehensive dataset featuring factors such as environmental conditions, socio-economic indicators, and historical wastage records, this study evaluates five prominent machine learning models: AdaBoost, XGBoost, CatBoost, Support Vector Machines (SVM), and Gaussian Naϊve Bayes. The models were assessed using metrics including Mean Squared Error, Root Mean Squared Error, R-Squared Score, and Mean Absolute Error. XGBoost emerged as the top performer, demonstrating superior predictive accuracy and robustness despite the inherent challenges. Additionally, feature importance analysis was conducted to identify significant factors influencing food wastage, offering valuable insights for stakeholders and policymakers. This study underscores the potential of advanced machine learning techniques in addressing the complexities of food waste prediction and advancing sustainability within the food distribution network.