A Time Series Study on the Growth of Fast Food Consumption and Its Impact on Public Health Indicators

This research examines the connections between the patterns of fast food consumption and the quality of health in the population based on data-driven and machine learning methods. The first goal will be to determine the relationship between increased fast food consumption and important health outcomes and also to test the predictive accuracy of various classifiers. Structured consumption and health data were subjected to statistical analysis, assessment of correlation and supervised learning models. The experimental findings indicate that consumption of fast foods has a strong association with negative health indicators and machine learning classifiers have competitiveness in terms of predicting the negative health outcomes. The results can be used to support the policy-level interventions and encourage the healthier eating habits. Moreover, the exploratory analysis of data and evidence showed significant patterns of dietary habits and behaviors which cause differences in health risk factors. Ensemble learning techniques, especially those of the random forest and the XGBoost, performed better than traditional classifiers and this demonstrates their ability to extract complex relationships in the data. The importance of features analysis showed that the frequency of fast food consumption was a prevalent predictor of adverse health effects. The suggested framework has shown the potential of machine learning models as tools to support the decision-making in the context of the evaluation of public health. Altogether, the experiment provides a comprehensive approach to dietary effects and the preventative health measures, which is more data-driven and scalable.