This entry explores a food pairing dataset that lists recipes by cuisine alongside their ingredient lists. The data comes from the supplementary material of a Nature Scientific Reports paper on flavour networks.

Loading the data

recipes_raw <- read_table(
  quarto_here(
    "srep00196-s3.csv"
  ),
  skip = 4,
  col_names = c("raw_recipe", "other")
)

Warning: 56498 parsing failures.
row col  expected    actual                                                                                      file
  1  -- 2 columns 1 columns '/Users/charliejhadley/Github/charliejhadley-website/posts/2022-04-20_2/srep00196-s3.csv'
  2  -- 2 columns 1 columns '/Users/charliejhadley/Github/charliejhadley-website/posts/2022-04-20_2/srep00196-s3.csv'
  3  -- 2 columns 1 columns '/Users/charliejhadley/Github/charliejhadley-website/posts/2022-04-20_2/srep00196-s3.csv'
  4  -- 2 columns 1 columns '/Users/charliejhadley/Github/charliejhadley-website/posts/2022-04-20_2/srep00196-s3.csv'
  5  -- 2 columns 1 columns '/Users/charliejhadley/Github/charliejhadley-website/posts/2022-04-20_2/srep00196-s3.csv'
... ... ......... ......... .........................................................................................
See problems(...) for more details.

recipes_data <- recipes_raw |>
  mutate(cuisine = str_extract(raw_recipe, "[A-z]{1,}(?=,)")) |>
  mutate(
    ingredients_list = str_remove(raw_recipe, "[A-z]{1,},"),
    ingredients_list = str_split(ingredients_list, ",")
  ) |>
  select(cuisine, ingredients_list) |>
  mutate(recipe_id = row_number()) |>
  relocate(recipe_id)

Most popular ingredients by cuisine

most_popular_ingr_by_cuisine <- recipes_data |>
  select(-recipe_id) |>
  unnest(ingredients_list) |>
  add_count(cuisine, ingredients_list, name = "n_ingredient_by_cuisine") |>
  unique() |>
  group_by(cuisine) |>
  slice_max(n_ingredient_by_cuisine, n = 10) |>
  ungroup()

most_popular_ingr_by_cuisine |>
  mutate(
    ingredients_list = reorder_within(
      ingredients_list,
      n_ingredient_by_cuisine,
      cuisine
    )
  ) |>
  ggplot(aes(y = n_ingredient_by_cuisine, x = ingredients_list)) +
  geom_col() +
  facet_wrap(~cuisine, scales = "free") +
  scale_x_reordered() +
  coord_flip() +
  labs(
    x = NULL,
    y = "Number of recipes",
    title = "Top 10 ingredients by cuisine"
  )

Scoring recipes by typicality

A recipe’s “typicality” score is the average popularity rank of its ingredients within its cuisine. Higher scores mean the recipe leans on the most-used ingredients for that style.

highest_scoring_recipes <- recipes_data |>
  unnest(ingredients_list) |>
  add_count(cuisine, ingredients_list, name = "n_ingredient_by_cuisine") |>
  group_by(recipe_id) |>
  mutate(score_ingredient_list = sum(n_ingredient_by_cuisine) / n()) |>
  ungroup() |>
  group_by(cuisine) |>
  slice_max(score_ingredient_list, n = 5) |>
  mutate(ingr = paste(ingredients_list, collapse = ", ")) |>
  ungroup() |>
  select(cuisine, score_ingredient_list, ingr)

highest_scoring_recipes |>
  knitr::kable(
    digits = 1,
    col.names = c("Cuisine", "Typicality score", "Ingredients")
  )

Cuisine	Typicality score	Ingredients
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
African	131.5	tomato, olive_oil, wheat, bell_pepper, garlic, cayenne
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EastAsian	1067.7	vinegar, soy_sauce, garlic, vinegar, soy_sauce, garlic
EasternEuropean	197.0	egg, wheat, butter, egg, wheat, cream
EasternEuropean	197.0	egg, wheat, butter, egg, wheat, cream
EasternEuropean	176.5	egg, wheat, butter, egg, wheat, cream
EasternEuropean	176.5	egg, wheat, butter, egg, wheat, cream
EasternEuropean	176.5	egg, wheat, butter, egg, wheat, cream
EasternEuropean	176.5	egg, wheat, butter, egg, wheat, cream
LatinAmerican	1762.5	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1762.5	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1762.5	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1762.5	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1609.2	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1609.2	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1609.2	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
LatinAmerican	1609.2	tomato, cayenne, tomato, cayenne, garlic, cayenne, cumin, onion
MiddleEastern	237.0	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	237.0	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
MiddleEastern	225.3	olive_oil, garlic, butter, egg, wheat, butter, egg, wheat, butter, egg, wheat
NorthAmerican	17088.0	butter, butter, butter, butter, butter
NorthAmerican	17088.0	butter, butter, butter, butter, butter
NorthAmerican	17088.0	butter, butter, butter, butter, butter
NorthAmerican	17088.0	butter, butter, butter, butter, butter
NorthAmerican	17088.0	butter, butter, butter, butter, butter
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
NorthernEuropean	145.7	butter, wheat, egg, butter, wheat, egg
SouthAsian	290.5	vegetable_oil, onion, vegetable_oil, cayenne, garlic
SouthAsian	290.5	vegetable_oil, onion, vegetable_oil, cayenne, garlic
SouthAsian	286.7	vegetable_oil, onion, vegetable_oil, cayenne, garlic
SouthAsian	286.7	vegetable_oil, onion, vegetable_oil, cayenne, garlic
SouthAsian	286.7	vegetable_oil, onion, vegetable_oil, cayenne, garlic
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SoutheastAsian	214.0	garlic, fish, lime_juice, garlic, fish, lime_juice
SouthernEuropean	2396.5	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	2396.5	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	2102.5	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	2102.5	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	1885.3	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	1885.3	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
SouthernEuropean	1885.3	olive_oil, garlic, tomato, olive_oil, tomato, olive_oil, onion
WesternEuropean	1395.0	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat
WesternEuropean	1336.5	butter, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat, butter, wheat