Examination and Evaluation of Obesity Risk Factors with Explainable Artificial Intelligence

There is an increasing need for effective methods for the detection and management of obesity,which is an important public health problem worldwide and is critical for the sustainability of health systems.This study examines the effectiveness of data preprocessing and machine learning techniques in detectingobesity. Data preprocessing steps, including the removal of unnecessary data, handling missing values, andaddressing data imbalance, are necessary to enhance the accuracy of machine learning algorithms. In thisstudy, data preprocessing steps were applied to an obesity dataset to make it suitable for machine learning.Using a dataset of 2111 patients, this study evaluates the effectiveness of machine learning techniquesin detecting obesity. Following the completion of data preprocessing, obesity was identified using variousmachine learning algorithms, including Decision Tree Classification, Random Forest Classification, NaiveBayes Classification, KNN, and XGBoosting, and their performances were compared. According to theresults, the XGBoosting algorithm exhibited the highest accuracy (0.92), precision, recall, and F1-scorevalues. Explainable Artificial Intelligence (XAI) techniques, such as SHAP and InterpretML, were employedto understand the effects of obesity parameters and determine which parameters have a greater impacton obesity. By visualizing and analyzing the effects of obesity parameters, these techniques facilitated theidentification of significant parameters in obesity detection. The findings demonstrate that the XGBoostingalgorithm outperforms other algorithms in detecting obesity. Furthermore, XAI techniques play a crucial role incomprehending obesity parameters. Specifically, a family history of obesity and factors like FCVC and CAECappear to have more significant effects compared to others