A comparison of a range of nutrient profiling models as applied to cooked meals
Bibliographic record
Abstract
Aim: A variety of different models for assessing the nutritive value of foods and ingredients exist, but all can be subject to criticism in not adequately assessing the full complement of nutrient components in a food or meal. This work was carried out in conjunction with an expert review panel comprised of technical experts from industry, academia, and policy, and aimed to compare some of the most prominent nutrient profiling methods when applied to a range of cooked meals. Method: Three different models were considered: the Ofcom Nutrient Profiling Model (NPM) which is based on a point system that includes 4 nutrients to discourage and 3 nutrients/food components to encourage; the Nutrient Rich Foods (NRF) Index, which ranks foods based on only their nutrient content (usually includes 9 nutrients to encourage and 3 to limit); and finally the SAIN,LIM model, which classifies foods into four healthiness classes based on separate indexes, the SAIN (which includes 5 nutrients to encourage) and LIM (which includes 3 nutrients to discourage). These models were applied to a variety of meals– meat based and vegetarian, rice based and pasta, dairy and dairy alternatives. Results: The Expert panel expressed a concern that the Ofcom NPM didn’t take micronutrients (e.g., iron, selenium) into account. However, Ofcom NPM considered food groups (i.e., percentage of fruit, vegetable and nut content) which will be micronutrient-rich and are regarded as a good way for testing if diets align with national dietary guidance. NRF9.3 was the most flexible index regarding the reference amount (e.g., 100grams, 100kcal and portion size). Comparing different cooked meals, vegetarian meals performed better than meat-based meals (e.g., spaghetti Bolognese made with beef versus Quorn); and meals that included dairy seemed to score more poorly than those which didn’t. Conclusion: Assessing the “healthiness” of meals using different methods is dependent on the choice of profiling model used. Scores should be considered under the lens of each model’s characteristics. Nutrient profiling models that are selected to influence consumers’ food choices and preferences should also align with both sustainability targets and nutritional guidelines.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.015 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.004 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".