Recipe Verification Based on Age-Adjusted Multiple Servings by JavaScript and R Language

Access to nutritious and healthy diets is crucial for children's health, growth, and development. The school meal programs play an integral role in ensuring that all children, regardless of their socio-economic background can receive high-quality and nutritious food. The primary goal of these programs is to reduce the occurrence of undernutrition, overweight, and obesity as well as to promote healthy diets. Although an extensive number of verified recipes are made available to help schools prepare healthy meals, many schools still prefer to use their recipes to make school meals. Unfortunately, most school-own recipes have not been verified, and therefore is difficult to guarantee the nutritional quality and outcome of recipes. In this paper, we presented a systematic approach to verify and correct the recipes using a combination of statistics and text analysis. The proposed method incorporated the age and number of servings to facilitate and improve the process of verification and correction of recipes. A total number of 7,505 school lunch recipes were used in the study. In particular, the dataset of 1,009 verified standardized recipes was trained using the proposed model to evaluate and revise the set of 6,496 unverified test data. The experimental results demonstrated the effectiveness and benefits of the proposed method for verifying the recipes. The results further illustrated that the presented approach suggested the revision of the invalid recipes.