Adult Obesity Prediction with Ensemble Classification Techniques - A Deep Learning Strategy

The development of precise prediction models for early intervention is of the utmost importance due to the significant public health issue of the increasing prevalence of adult obesity. The study introduces an innovative approach for adult obesity prediction using deep learning and ensemble classification techniques. The current proposed approach is trying to overcome previous approaches in terms of robustness and accuracy. Normally, the proposed method is improving the prediction in performance by collecting various classifiers into one framework based on deep learning. Several classifiers are considered here, such as decision trees, SVMs, k-nearest neighbors. The system employs dynamic weighting modification scheme that optimizes input from every classifier within the training phase. On such a large dataset, experiments indicate that the proposed model is much more reliable and accurate compared to standard single-model methods for adult obesity prediction. These results showed that public health professionals may benefit from the ensemble learning that improves obesity prediction with a potential accuracy of 86.7<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> in monitoring populations and responding promptly, when necessary.