The journal’s Editorial Office and Editorial Board are jointly issuing a resolution and removal of the Journal Notice linked to this article [1], as well as an update to the original publication. Following concerns raised about the integrity of the peer review, the Editorial Office has conducted a post-publication peer review of this article. This process included the recruitment of one new independent reviewer and was supervised by the Editorial Board member and the Editors-in-Chief to ensure full compliance with MDPI’s editorial process (https://www.mdpi.com/editorial_process, accessed on 5 March 2026).
As a result of this process, the Editors-in-Chief and the authors have agreed to update the following aspects of this publication:
Reviewer report 2 was removed from the peer review record, and one new report was uploaded to the peer review record (https://www.mdpi.com/2072-6643/13/9/3195/review_report, accessed on 30 March 2026).
Text Correction
In the original publication [1], we updated the contents in the Materials and Methods Section. With this update, we clarified that the nutrition data was obtained from nutrition information panels. We also expanded our rationale for the exclusion criteria and the chosen train–test split.
In Section 2.1, the second-to-last sentence was newly added: “All nutrition information was derived from the nutrition information panel.”.
In Section 2.2, in the first paragraph, the second-to-last sentence was updated to: “Additionally, we excluded food and beverage categories that either (i) contributed less than 0.8% of fiber intake in Australia as this cutoff represented the lower tail of the distribution of fiber contribution across categories [30] (e.g., ‘cheese’, ‘cooking oils’, and ‘honey’), or (ii) had less than 25% of products reporting fiber content (e.g., ‘processed meat’) (n = 41,338) (Supplementary Table S1).”.
In Section 2.2, the second paragraph, the third sentence was newly added: “Stratifying by brand name allowed us to avoid train-test leakage due to near-duplicate products (e.g., it would be inappropriate to train on “Pringles Original” and test on “Pringles BBQ”, as it is unlikely that a brand would report fiber content on one product and not the other in the real-world); however, it did slightly disrupt the proportion of different food categories between the training and test dataset (i.e., the proportion of all products in the test dataset ranged from 11% for ‘pasta’ and 34% for ‘biscuits’).”.
Data Availability Statement Updated
A correction has been made to the Data Availability Statement. With this update, we have provided the code used for the machine learning algorithm.
The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Reference
- 1.Davies T., Louie J.C.Y., Scapin T., Pettigrew S., Wu J.H., Marklund M., Coyle D.H. An Innovative Machine Learning Approach to Predict the Dietary Fiber Content of Packaged Foods. Nutrients. 2021;13:3195. doi: 10.3390/nu13093195. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
A correction has been made to the Data Availability Statement. With this update, we have provided the code used for the machine learning algorithm.
The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.