Data quality error: category and computed Nutri-Score is not coherent

[CharlesNepote]

The Nutri-Score of some categories is predictable. Eg:

• Extra-virgin olive oils => Nutri-Score C
• Apples => A
• etc.

If an extra-virgin olive oils is not Nutri-Score C, then:

• either the category is wrong
• either the nutrition facts are wrong.

In both cases, it's an error, there can't be false positives. It should lead to a data-quality-error, and not just a warning.

It probably represent thousands of products. For example, as of today (2023-04), 677 olive oils have a Nutri-Score different from C.

To let it scale, I would be in favor to add this data in the taxonomy. Eg. (look at the last line):

<en:Virgin olive oils
en:Extra-virgin olive oils, Extra virgin olive oil
bg:Зехтин екстра върджин
de:Native Olivenöle Extra, Natives Olivenöl Extra
fr:Huiles d'olive vierges extra, Huile d'olive vierge extra
agribalyse_food_code:en:17270
ciqual_food_code:en:17270
ciqual_food_name:en:Olive oil, extra virgin
ciqual_food_name:fr:Huile d'olive vierge extra
expected_nutriscore:en:c
expected_ingredients_text:en:Olive oil

Part of

• 🚦Nutri-Score (tracker) #5517
• 🧽 Data quality (tracker) #5538

[benbenben2]

@CharlesNepote, is it done?

[CharlesNepote]

I guess this is not done yet. I have just modified this olive oil, but it's still Nutri-Score D and without data quality error: https://world.openfoodfacts.org/product/001682/minotaur-bio-huile-d-olive-vierge-extra
Maybe @stephanegigandet has to launch taxonomy deployment or some task I don't know of.

[benbenben2]

DQ errors are there now @CharlesNepote

[CharlesNepote]

Yes. For the record:

• https://world.openfoodfacts.org/data-quality-error/en:nutri-score-grade-from-category-does-not-match-calculated-grade (95 products as of 2023-09-11)
• https://world.openfoodfacts.org/data-quality-error/en:ingredients-single-ingredient-from-category-does-not-match-actual-ingredients (50 products as of 2023-09-11)

Also added to: https://wiki.openfoodfacts.org/List_of_data_quality_errors_(generated)

Before closing I'm going to open another issue to gather more products with only one ingredient, and/or product categories which should have only one Nutri-Score.

[github-actions]

This issue has been open 90 days with no activity. Can you give it a little love by linking it to a parent issue, adding relevant labels and projets, creating a mockup if applicable, adding code pointers from https://github.com/openfoodfacts/openfoodfacts-server/blob/main/.github/labeler.yml, giving it a priority, editing the original issue to have a more comprehensive description… Thank you very much for your contribution to 🍊 Open Food Facts

[benbenben2]

@CharlesNepote, is there anything else to do with this issue?

[CharlesNepote]

Stats for 2024-09-02:

• https://world.openfoodfacts.org/data-quality-error/en:nutri-score-grade-from-category-does-not-match-calculated-grade => 83 products (-10% for one year)
• https://world.openfoodfacts.org/data-quality-error/en:ingredients-single-ingredient-from-category-does-not-match-actual-ingredients => 689 products (+1000% for one year)

In most case this is due to olive oil which should always contain only "olive oil" but which often countain "olive oil manually blah blah blah".

Should en:ingredients-single-ingredient-from-category-does-not-match-actual-ingredients be an error or a warning?