This research explores the application of machine learning techniques, specifically regression models, for predicting the nutrient composition of various food items. The dataset utilized in this study, referred to as 'food.csv,' includes a wide range of nutritional information, such as vitamin, mineral, protein, carbohydrate, and fat content, alongside household weight data for specific food items. The objective was to develop a model capable of predicting the nutritional content from these attributes. After preprocessing the data and splitting it into training and testing sets, we applied a regression model and evaluated its performance using several metrics, including Mean Squared Error (MSE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R² score. The results revealed that the regression model was highly accurate, achieving an R² score of 0.99, which indicates that the model was able to explain 99% of the variance in the nutritional data. The successful application of this model demonstrates its potential utility in areas such as personalized nutrition, food industry optimization, and public health research, where accurate predictions of nutrient content are essential.