Hugging Face's logo

Datasets:
devnarsinghani
/
india-label-index

Modalities:
Tabular
Text
Formats:
csv
Languages:
English
Size:
1K - 10K
Tags:
india
food
nutrition
food-labels
fssai
open-food-facts
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
Dataset Viewer
Auto-converted to Parquet Duplicate
category
stringlengths
14
36
products
int64
6
792
with_ingredient_list
int64
1
344
pct_any_added_sugar
float64
0.9
100
pct_sugar_in_top3
float64
0.9
100
pct_2plus_sugar_names
float64
0.9
100
pct_3plus_sugar_names
float64
0
34.3
max_sugar_names
int64
2
5
with_sugar_value
int64
2
103
median_sugar_g_per_100g
float64
0
46.3
median_sugar_tsp_per_100g
float64
0
11.6
with_salt_value
int64
2
105
median_salt_g_per_100g
float64
0
2.5
median_salt_pct_of_5g
int64
0
50
with_nova
int64
3
435
pct_nova4
float64
8.8
100
Biscuits and cookies
792
344
93
80.2
70.1
34.3
5
96
25
6.2
90
0.57
11
365
94.2
Sauces, ketchup and spreads
719
337
66.2
55.2
6.8
0.9
3
99
9
2.2
93
0.34
7
435
46.4
Flours, grains and pulses
623
303
8.6
3.3
1.7
0.7
3
56
1.54
0.4
55
0.06
1
302
13.2
Chocolate and confectionery
496
236
86
77.1
34.3
14.8
4
82
46.33
11.6
77
0.24
5
235
92.8
Spices and masalas
446
199
12.1
3
1
0
2
22
2.15
0.5
24
1.62
32
203
17.2
Dairy (milk, paneer, cheese)
393
221
16.7
16.3
3.2
0.5
3
61
4.7
1.2
62
0.13
3
239
25.9
Cakes and bakery
333
166
86.1
50
10.8
0.6
3
38
3.81
1
36
1.07
21
169
76.3
Chips and crisps
308
163
65.6
16
11.7
0
2
60
2.5
0.6
58
1.69
34
165
71.5
Tea and coffee
273
66
16.7
13.6
3
0
2
19
0
0
19
0
0
83
25.3
Breakfast cereals and muesli
256
150
64
33.3
36
19.3
5
56
9.88
2.5
54
0.25
5
148
56.8
Instant noodles and pasta
251
116
67.2
2.6
4.3
0
2
49
2.2
0.6
43
2.5
50
131
87
Soft drinks and energy drinks
241
130
74.6
73.1
11.5
1.5
3
60
10.55
2.6
57
0.02
0
141
93.6
Edible oils and ghee
235
107
0.9
0.9
0.9
0
2
22
0
0
23
0
0
205
8.8
Milk drinks and health drink powders
193
112
52.7
51.8
11.6
4.5
4
23
16.11
4
24
0.14
3
105
60
Yoghurt and curd
190
111
30.6
23.4
5.4
0
2
103
5
1.2
105
0.12
2
92
48.9
Ice cream and frozen desserts
178
106
86.8
70.8
43.4
12.3
4
25
22.5
5.6
23
0.12
2
104
95.2
Namkeen and savoury snacks
173
105
52.4
9.5
6.7
1.9
4
30
3.4
0.8
25
1.42
28
101
47.5
Fruit juices and juice drinks
152
91
84.6
82.4
29.7
4.4
3
25
11.84
3
22
0.05
1
91
84.6
Ready meals and instant mixes
78
27
66.7
37
18.5
0
2
17
3.3
0.8
17
1.2
24
32
78.1
Protein and snack bars
6
1
100
100
100
0
2
2
27.3
6.8
2
0.65
13
3
100

India Label Index

This dataset is a collection of category-level tables from a study of Indian packaged food labels on Open Food Facts. It includes counts of added sugars, fats, warnings, and other nutritional data. The data is useful for analyzing food labeling trends and nutritional content in India.

What is inside

The dataset contains 13 files derived from the India Label Index, compiled using Open Food Facts products available in India. These include category-level tables such as category_table.csv and results.json, which provide data on added sugar, sugar in the first three ingredients, salt, and NOVA 4 share by category, based on 3,798 non-staple ingredient lists. Additional files like warnings.json and fssai.json detail Chile’s 2019 warning-label rule and FSSAI’s proposed rule applied to Indian foods. Files such as fats.json, world.json, sugar_world.json, names.json, and diwali.json offer insights into fat types, international comparisons, naming conventions, and festive food products. Each share is recorded with its count (k) and base (n).

File Size Script Study page
category_table.csv 1.9 KB analyse.py https://devnarsinghani22.github.io/india-label-index/
results.json 19.2 KB analyse.py https://devnarsinghani22.github.io/india-label-index/
warnings.json 3.1 KB warnings.py https://devnarsinghani22.github.io/india-label-index/warning-labels/
fssai.json 2.9 KB fssai.py https://devnarsinghani22.github.io/india-label-index/fssai-warning-labels/
fats.json 8.3 KB fats.py https://devnarsinghani22.github.io/india-label-index/palm-oil/
world.json 23.5 KB world.py https://devnarsinghani22.github.io/india-label-index/india-vs-world/
world_table.csv 13.4 KB world.py https://devnarsinghani22.github.io/india-label-index/india-vs-world/
sugar_world.json 97.9 KB sugar_world.py https://devnarsinghani22.github.io/india-label-index/sugar-salt-world/
sugar_world_table.csv 45.4 KB sugar_world.py https://devnarsinghani22.github.io/india-label-index/sugar-salt-world/
names.json 2.6 KB names.py https://devnarsinghani22.github.io/india-label-index/healthy-names/
names_table.csv 0.3 KB names.py https://devnarsinghani22.github.io/india-label-index/healthy-names/
diwali.json 9.1 KB diwali.py https://devnarsinghani22.github.io/india-label-index/diwali/
diwali_table.csv 3.1 KB diwali.py https://devnarsinghani22.github.io/india-label-index/diwali/

Findings

  • 60% of 3,798 Indian non-staple ingredient lists name an added sugar; 42% have a sugar in the first 3 ingredients; 19% list 2 or more sugar names (biscuits 70%); 68% are ultra-processed (NOVA 4).
  • Instant noodles: median 2.5 g salt per 100 g.
  • Under Chile's warning rule, 69% of 687 Indian foods get at least 1 warning and 53% get 2 or more; biscuits 98%, chocolate 96%, chips 94%.
  • Under FSSAI's proposed rule (any 1 nutrient high), 71% of 665 non-staple foods would carry a warning; under the dropped 2-nutrient plan, 41%. Soft drinks: 69% under the current plan, 3% under the dropped plan.
  • 52% of 2,015 lists naming a fat name palm oil; instant noodles 90%, biscuits 82%, chips 77%, namkeen 66%. 15% name a hydrogenated or interesterified fat.
  • Indian chips: 78% (124 of 159) list palm oil, against 3.3% (46 of 1,380) in the UK and 7.6% (211 of 2,779) in the US.
  • Biscuits with 5 or more additives: 45% (158 of 350) in India, the most of 12 countries.
  • Grain-word packs: 21 of 27 warned under FSSAI's proposed rule, against 403 of 482 plain packs.

Method

The data comes from the Open Food Facts CSV export, with rows tagged as sold in India (pulled 1 October 2026). A 16-country pull was done on 6 October 2026 for the world files. Baby food and supplements were left out. Staples were left out of the India base. Sugar names were counted using an alias list (sugar, invert syrup, glucose syrup, jaggery, dextrose, maltodextrin, etc.). Fats were counted using a matcher for palm, named plant oils, dairy fats, and hydrogenated fats. Warnings were applied by checking each rule's per 100g or per 100ml limits against the label's declared values, counting a nutrient only when it was added. Every number is rounded once from k and n. Scripts in the GitHub repo rebuild every file.

Caveats

  • The lists come from Open Food Facts, entered by volunteers; they are not a random sample of Indian shops, so the shares describe this collection, not the market.
  • FSSAI's warning label is a proposal; the Supreme Court has not ruled on it.
  • The fat count is an upper estimate.
  • No brands are named; the files cannot be used to rank or name a brand.
  • Where k is small, read the shares as rough.
  • Ingredient text was matched by pattern, so a misspelt or missing list is not counted.

Citation

Ask readers to cite the DOI 10.5281/zenodo.23245061 and to credit Open Food Facts (ODbL).

Narsinghani, D. (2026). India Label Index: category-level study of Indian packaged-food labels from Open Food Facts. Zenodo. https://doi.org/10.5281/zenodo.23245061
Downloads last month
75
Number of rows:
1,051
Total file size:
245 kB