Data and code underlying the publication: Using diet optimization and machine learning for the design of healthy and acceptable menu plans

The objective of this study was to contribute to the development of data-driven consumer acceptance modelling by integrating diet optimization with recipe completion, and comparing this approach to traditional diet modelling methods, which rely on food group filtering and item popularity. For the design of healthy and acceptable vegetarian menu plans, We used recipe completion to identify substitutes that go well together with the remaining food items within a meal and diet modelling to replace food items with healthier alternatives. Please view https://doi.org/10.1016/j.ejor.2025.06.015 for added information on materials and methods. - Consumption dataVegetarian meals that served as model input were created based on NHANES consumption data. Food group equivalents data (FPED) and the WWEIA and FNDDS food group classifications were used to label meals as vegetarian or not. The input data is further described in FolderContents - 1 Raw data. The processing of the data was done in R, see FolderContents - 2 Model input data. - Recipe completion modelLinear kernel ridge regression was applied to quantify the compatibility of potential substitutes with other food items within a meal (see FolderContents - 3 Recipe completion model). - Diet modelTwo diet models were developed to improve the healthiness of meals by substituting food items. Our approach replaced food items using the recipe completion algorithm, while the other model followed the traditional approach by selecting substitutes based on food group filtering and food popularity (see FolderContents - 4 Diet model). - Processing and analysis scriptsAll processing of data (modifications, calculations, etc.) and analysis of data was performed in R. All scripts are elaborately commented, describing each different step taken and the reason for the step. - Figures and tablesAll figures and tables were created through R. Refer to the R scripts for detailed information within the script.