Individuals with health conditions and special preferences—especially those of the senior population—often have a hard time cooking and preparing healthy meals for themselves, mainly because of lack of knowledge regarding their specific diet like diabetes and high blood pressure. Personalized nutrition plays a crucial role in promoting a healthier lifestyle and minimizing food waste, making it increasingly important to accurately understand the nutritional composition of foods. Most current individualized nutrition assistants use lookup tables for these values, which are sometimes inconvenient for users with uncommon names of foods. This research project leverages a dataset of 2,395 names of foods with precise macronutrient and vitamin data to train a Machine Learning model. The model predicts numerical nutritional values such as carbohydrates, sugar, and saturated fat, based on the name of the food (given 100g). This project will optimize the performance of the model by using different training methods, including custom Neural Network models and Large Language models, and analyzing their accuracies. By analyzing the names’ semantic meanings, this research will be able to tackle this problem and advance personalized nutrition solutions.