Exploring and Modeling Factors Influencing Obesity with a Comprehensive Analysis Using Synthetic and Real-World Data

The current investigation focuses on the necessary examination of both generated data and actual data regarding obesity. Analyzing a dataset from Kaggle, the study compares demographic, lifestyle, and health characteristics associated with obesity. Steps such as data cleaning, data feature extraction and selection, and machine learning algorithms including Decision Trees, Random Forest, Neural Networks were also used in making the predictions of risk for obesity. Findings in the study show that results from application of Neural Networks among all models are superior; this methods of analysis offer high accuracy in risk-based subject identification. The feature selection analysis shows that such aspects as age, physical activity, and diet has a greater impact. Statistical results and model differencing produce features about obesity risk on potentially useful in public health activities. The work describes a contribution to the field of health informatics in the application of predictive modeling and is a framework for future investigations into prevention of obesity among various individuals.

Перевод пока недоступен