Difference between revisions of "Ingredients Analysis Quality Evaluation - June 2022"

From Open Food Facts wiki
Jump to navigation Jump to search
(Record current state of ingredients analysis for IN, UK, US, to track improvements)
(No difference)

Revision as of 07:40, 23 June 2022

Evaluation of the quality of Ingredients Extraction and Analysis so that we can measure the improvements.

Ingredients parsing and recognition

For each product, the ingredients list is parsed to separate each ingredient. Ingredients that we can match to our multilingual ingredients taxonomy as marked as "known", others as "unknown".

There are different reasons an ingredient can be marked as unknown:

  • The input ingredient list is incorrect (misspellings etc.)
  • The ingredients list contains other things than ingredients and we have not been able to split it at the right place.
  • We have not been able to parse a particular sentence structure or formatting (e.g. "a drop of delicious honey with a shower of powder sugar")
  • The ingredient is not yet present in our taxonomy (or not with the right synonym or in the right language)

In the table below,

  • the first column of numbers corresponds to the number of unique ingredients across all products.
  • And the second column of number corresponds to the number of occurrences of those ingredients. e.g. if a specific ingredient appears in 5 products, it is counted 5 times.


EN - English - India

2022-06-23
All products:
known	772 (22.89%)	8300 (69.16%)
unknown	2600 (77.11%)	3702 (30.84%)
all	3372 (100.00%)	12002 (100.00%)

Top 10k most scanned products:
known	408 (35.73%)	2102 (68.63%)
unknown	734 (64.27%)	961 (31.37%)
all	1142 (100.00%)	3063 (100.00%)

EN - English - UK

2022-06-23
All products:
known	2591 (6.23%)	406395 (88.32%)
unknown	39005 (93.77%)	53757 (11.68%)
all	41596 (100.00%)	460152 (100.00%)

Top 10k most scanned products:
known	1664 (21.70%)	77611 (91.15%)
unknown	6003 (78.30%)	7539 (8.85%)
all	7667 (100.00%)	85150 (100.00%)


EN - English - UK

2022-06-23
All products:
known	3027 (1.77%)	8121096 (91.22%)
unknown	167711 (98.23%)	781824 (8.78%)
all	170738 (100.00%)	8902920 (100.00%)

Top 10k most scanned products:
known	1619 (18.25%)	97759 (88.84%)
unknown	7252 (81.75%)	12280 (11.16%)
all	8871 (100.00%)	110039 (100.00%)

Ingredients analysis

For each product, once we have parsed the ingredients list to separate each ingredient and match them to our taxonomy, we can analyze if the products are vegan, vegetarian, without palm oil etc.

To mark a product as non-vegan etc. we only need to find one non-vegan ingredient. But to mark a product as vegan, we need to be sure that we have been able to correctly recognize all ingredients so that we can verify that all of them are vegan.

EN - English - IN

2022-06-23
All products:
Vegetarian status unknown	478	
Palm oil content unknown	436	
Vegan status unknown	390	
Non-vegan	184	
Vegetarian	159	
Palm oil free	96	
Vegan	79	
Palm oil	70	
May contain palm oil	49	
Non-vegetarian	20	
Maybe vegan	8	
Maybe vegetarian	7		
Top 10k most scanned products:
Vegetarian status unknown	128	
Palm oil content unknown	114	
Vegan status unknown	101	
Non-vegan	43	
Vegetarian	26	
Palm oil	22	
Palm oil free	16	
Vegan	12	
May contain palm oil	5	
Non-vegetarian	3	
Maybe vegan	2	
Maybe vegetarian	2		

EN - English - UK

2022-06-23
All products:
Vegetarian status unknown	8623	
Palm oil content unknown	8378	
Non-vegan	7693	
Vegetarian	7293	
Vegan status unknown	6188	
Palm oil free	5725	
Vegan	5021	
Non-vegetarian	2754	
Palm oil	2660	
May contain palm oil	1648	
Maybe vegetarian	911	
Maybe vegan	559	
Top 10k most scanned products:
Vegetarian	1741	
Palm oil content unknown	1472	
Vegetarian status unknown	1412	
Palm oil free	1407	
Non-vegan	1359	
Vegan	1138	
Vegan status unknown	1120	
Palm oil	511	
Non-vegetarian	434	
May contain palm oil	270	
Maybe vegetarian	159	
Maybe vegan	112	

EN - English - US

2022-06-23
All products:
Vegetarian status unknown	212861	
Palm oil content unknown	176245	
Non-vegan	146466	
Vegan status unknown	126220	
Palm oil free	85478	
Vegetarian	51116	
Non-vegetarian	45193	
Vegan	44942	
Palm oil	37921	
May contain palm oil	30916	
Maybe vegetarian	21742	
Maybe vegan	13279	
Top 10k most scanned products:
Vegetarian status unknown	2914	
Palm oil content unknown	2488	
Vegan status unknown	1954	
Non-vegan	1552	
Palm oil free	1223	
Vegetarian	944	
Vegan	847	
Palm oil	405	
Non-vegetarian	391	
May contain palm oil	361	
Maybe vegetarian	248	
Maybe vegan	143		


Observations