A DEEP LEARNING FRAMEWORK FOR PREDICTIVE MODELING OF ADOLESCENT OBESITY
Keywords:
adolescent obesity, deep learning, predictive modeling, artificial neural networks (ANN), convolutional neural networks (CNN), long short-term memory (LSTM)Abstract
This paper endeavors to identify the underlying causes of adolescent obesity and promote early intervention by utilising sophisticated deep learning algorithms to predict its growth. The issue of adolescent obesity is causing growing concern in the global public health sector. This is the result of genetics, insufficient exercise, unhealthy eating practices, and excessive mental stress. This research encourages the utilization of deep learning techniques, such as ANN, CNN, and LSTM models, to analyse extensive health datasets that include demographic, behavioral, clinical, and lifestyle variables. The program's objective is to accurately predict the prevalence of obesity among adolescents by evaluating a substantial volume of intricate, high-dimensional data and classifying individuals according to their likelihood of becoming overweight. The paper's primary objective is to evaluate efficacy through the use of metrics such as F1-score, recall, accuracy, and precision, in addition to feature extraction, data purification, and model training. The proposed method is expected to enhance the precision of predictions, facilitate physician decision-making, and facilitate the development of intelligent healthcare systems for the prevention and management of adolescent obesity.
