{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:04:54Z","timestamp":1789621494622,"version":"build-2803163510"},"reference-count":16,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T00:00:00Z","timestamp":1789603200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Nutr."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Nutrient profiles contain information associated with food processing, but estimates of automated NOVA classification can be distorted by ambiguous record linkage, severe class imbalance, duplicate profiles, and non-nested model selection. This study evaluates how well four-class NOVA labels can be recovered from nutrient composition after addressing these sources of optimism.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The analysis used 2,484 manually labeled foods from the USDA Food and Nutrient Database for Dietary Studies, represented by 99 nutrient variables. Identical nutrient profiles were assigned to the same fold, and logistic regression, a radial-basis support vector machine, random forest, and histogram gradient boosting were compared with a prior-probability baseline using three repeats of five-fold nested stratified group cross-validation. Hyperparameters and all transformations were selected within each outer training partition. Performance was summarized using class-balanced and chance-corrected metrics, corrected repeated-cross-validation tests, and fold-stable multiclass Shapley attribution.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Histogram gradient boosting achieved accuracy 0.921 and macro-F1 0.814 (SD 0.029; 95% fold interval 0.798\u20130.831). Random forest reached macro-F1 0.788; the mean difference of 0.026 was not statistically reliable after repeated-fold correction (Holm-adjusted\n                      <jats:italic>p<\/jats:italic>\n                      = 0.203). Class-specific F1 was 0.639 for NOVA 2 and 0.958 for NOVA 4.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Nutrient measurements therefore provide a strong but incomplete correlate of NOVA class. The models are suitable for methodological screening within this data source, but they do not replace ingredient- and process-based expert classification; external validation on contemporary branded foods and independently adjudicated labels remains necessary.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnut.2026.1916576","type":"journal-article","created":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T04:47:17Z","timestamp":1789620437000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Robustness of nutrient-based NOVA food classification under profile-separated nested validation"],"prefix":"10.3389","volume":"13","author":[{"given":"S\u0131la Sinem","family":"G\u00f6ren","sequence":"first","affiliation":[{"name":"Department of Business, University of Europe for Applied Sciences","place":["Potsdam, Brandenburg, Germany"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Raja Hashim","family":"Ali","sequence":"additional","affiliation":[{"name":"Department of Business, University of Europe for Applied Sciences","place":["Potsdam, Brandenburg, Germany"]},{"name":"Department of Artificial Intelligence, Ghulam Ishaq Khan Institute of Engineering Sciences and Technology","place":["Topi, Pakistan"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Talha Ali","family":"Khan","sequence":"additional","affiliation":[{"name":"Department of Business, University of Europe for Applied Sciences","place":["Potsdam, Brandenburg, Germany"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Iftikhar","family":"Ahmed","sequence":"additional","affiliation":[{"name":"Department of Business, University of Europe for Applied Sciences","place":["Potsdam, Brandenburg, Germany"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,17]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1017\/S1368980017000234","article-title":"The UN decade of nutrition, the NOVA food classification and the trouble with ultra-processing","volume":"21","author":"Monteiro","year":"2018","journal-title":"Public Health Nutr"},{"key":"B2","doi-asserted-by":"publisher","first-page":"936","DOI":"10.1017\/S1368980018003762","article-title":"Ultra-processed foods: what they are and how to identify them","volume":"22","author":"Monteiro","year":"2019","journal-title":"Public Health Nutr"},{"key":"B3","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1016\/j.tifs.2021.02.059","article-title":"Sokolovi\u0107 M, Timotijevic L. Processed food classification: conceptualisation and challenges","volume":"112","author":"Sadler","year":"2021","journal-title":"Trends Food Sci Technol"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1245","DOI":"10.1038\/s41430-022-01099-1","article-title":"Ultra-processed foods: how functional is the NOVA system?","volume":"76","author":"Braesco","year":"2022","journal-title":"Eur J Clin Nutr"},{"key":"B5","doi-asserted-by":"publisher","first-page":"2312","DOI":"10.1038\/s41467-023-37457-1","article-title":"Machine learning prediction of the degree of food processing","volume":"14","author":"Menichetti","year":"2023","journal-title":"Nat Commun"},{"key":"B6","doi-asserted-by":"publisher","first-page":"70","DOI":"10.3389\/fnut.2019.00070","article-title":"Characterizing ultra-processed foods by energy density, nutrient density, and cost","volume":"6","author":"Gupta","year":"2019","journal-title":"Front Nutr"},{"key":"B7","doi-asserted-by":"publisher","first-page":"103848","DOI":"10.1016\/j.jfca.2021.103848","article-title":"A study of ultra-processing marker profiles in 22,028 packaged ultra-processed foods using the siga classification","volume":"99","author":"Davidou","year":"2021","journal-title":"J Food Compos Anal"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.7160\/aol.2024.160207","article-title":"Analysing household food consumption in Turkey using machine learning techniques","author":"\u00d6ztornac\u0131B","year":"2024","journal-title":"AGRIS on-line Pap Econ Inform"},{"key":"B9","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1186\/1471-2105-7-91","article-title":"Bias in error estimation when using cross-validation for model selection","volume":"7","author":"Varma","year":"2006","journal-title":"BMC Bioinformatics"},{"key":"B10","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1023\/A:1024068626366","article-title":"Inference for the generalization error","volume":"52","author":"Nadeau","year":"2003","journal-title":"Mach Learn"},{"key":"B11","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: a gradient boosting machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann Stat"},{"key":"B13","doi-asserted-by":"publisher","first-page":"307","DOI":"10.1515\/9781400881970-018","article-title":"A Value for n-Person Games","author":"Shapley","year":"1953","journal-title":"Contributions to the Theory of Games II"},{"key":"B14","first-page":"4765","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"B15","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"B16","unstructured":"FoodData Central\n          \n          2026"}],"container-title":["Frontiers in Nutrition"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnut.2026.1916576\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T04:47:22Z","timestamp":1789620442000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnut.2026.1916576\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,17]]},"references-count":16,"alternative-id":["10.3389\/fnut.2026.1916576"],"URL":"https:\/\/doi.org\/10.3389\/fnut.2026.1916576","relation":{},"ISSN":["2296-861X"],"issn-type":[{"value":"2296-861X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,17]]},"article-number":"1916576"}}