Assessing and Prediction of the Nutritional Value of Fast-Food Meals: Calories and Macronutrients

This research is a comprehensive exploration into the nutritional aspects of fast-food meals in prominent U.S. restaurant chains, utilizing advanced machine learning techniques to predict calorie content and classify meals. The study employs various models, including Linear Regression, Decision Tree, Random Forest, and KNN for calorie prediction, as well as Random Forest and Naïve Bayes for meal classification. The analysis goes beyond calorie prediction and delves into the nuanced relationship between calories and nutrient profiles, offering a detailed panorama of nutrient values associated with consumer-purchased fast-food meals. The study takes a holistic approach by incorporating clustering methods such as K-mean and hierarchical clustering, providing insights into macronutrient profiles within each cluster. Key findings of the research point out Linear Regression as the most accurate predictor, achieving a remarkable 98.6% R-squared value and minimal mean squared error. This emphasizes its effectiveness in predicting calorie content in fast-food meals. Moreover, the study identifies total fat, carbohydrates, and protein as pivotal factors influencing calorie predictions. These findings contribute valuable insights for individuals seeking to make informed dietary decisions in the complex landscape of fast-food consumption. Overall, the research provides a comprehensive understanding of the nutritional intricacies of fast-food meals, combining machine learning techniques and clustering methods to reveal essential insights for consumers, researchers, and the food industry.