Author ORCID Identifier
0000-0003-3160-2058
Advisor Information
Kathryn Cooper
Location
Poster number 510 MBSC Ballroom
Presentation Type
Poster
Start Date
4-3-2022 12:00 AM
End Date
4-3-2022 12:00 AM
Abstract
A food label provides some of the most crucial information for a food product. The food label is a key resource for many health-conscious consumers for understanding ingredients. It is also vital for individuals to avoid food allergens or help patients follow dietary recommendations. While the food labels in the United States are regulated by the Food and Drug Administration (FDA) many labels contain additional information or statements that are not regulated. Moreover, the food label may be complex or contain terminology that the layperson may not understand. Evidence has indicated that consumers often find nutrition labels confusing, especially when it comes to more technical terms, such as ingredients used for fortification, emulsifiers, or additives. As a result, there are ongoing efforts to develop label designs and strategies that help consumer understanding. This work presents an ingredient label classification system using FoodOn, an ontology for food to categorize terminology found on food labels. First, food labels are tokenized by removing stop words and punctuations, leaving only the ingredients. All the ingredients are then searched for in the Ontology to classify. The results list ingredients found on food products classified into broader groups. Some examples are milled grain, additive, or refined products. This work showcases the feasibility of a system to classify ingredients which can make it easier to understand the origin of less commonly understood terms found on food labels.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Scheduling Link
1
Included in
Bioinformatics Commons, Data Science Commons, Dietetics and Clinical Nutrition Commons, Other Nutrition Commons, Public Health Education and Promotion Commons
Ingredient Classification Using Food Ontology
Poster number 510 MBSC Ballroom
A food label provides some of the most crucial information for a food product. The food label is a key resource for many health-conscious consumers for understanding ingredients. It is also vital for individuals to avoid food allergens or help patients follow dietary recommendations. While the food labels in the United States are regulated by the Food and Drug Administration (FDA) many labels contain additional information or statements that are not regulated. Moreover, the food label may be complex or contain terminology that the layperson may not understand. Evidence has indicated that consumers often find nutrition labels confusing, especially when it comes to more technical terms, such as ingredients used for fortification, emulsifiers, or additives. As a result, there are ongoing efforts to develop label designs and strategies that help consumer understanding. This work presents an ingredient label classification system using FoodOn, an ontology for food to categorize terminology found on food labels. First, food labels are tokenized by removing stop words and punctuations, leaving only the ingredients. All the ingredients are then searched for in the Ontology to classify. The results list ingredients found on food products classified into broader groups. Some examples are milled grain, additive, or refined products. This work showcases the feasibility of a system to classify ingredients which can make it easier to understand the origin of less commonly understood terms found on food labels.