CookBook-A Machine Learning-based Nutritional Analysis System Leveraging SVM for Personalized Diet Planning

Unhealthy meal plans have greatly aided in the upsurge of diet-related diseases like diabetes, hypertension, and obesity. Conventional meal planning seldom takes into account the health conditions of individuals, leading to a lot of generalized diets, devoid of personalization. The absence of such a tailored approach to dietary licensing runs the risk of noncompliance while increasing health hazards. To counter this, CookBook-a web framework-has been built to deliver intelligent health-based meal recommendations, powered by machine learning. Users can parametrize CookBook by entering their health parameters like blood sugar level, blood pressure, allergies, and dietary preferences to create a personalized nutrition profile. CookBook utilizes the support vector machine (SVM) algorithm, therefore providing an intelligent classification and recommendation through nutrient composition and individual’s health constraints. In this method, an accuracy of 96% was obtained, thus giving a better classification than any of its counterparts, such as logistic regression, decision tree, and random forests. Using AI-powered nutritional intelligence, CookBook improves compliance with dietary prescriptions, promotes healthier eating behavior, and offers personalized meal planning; therefore, it can be a very powerful tool for public health intervention.