Measured pairing · from 63,696 recipes

What goes with soybean?

Measured across 63,696 recipes, the 8 ingredients that co-occur with soybean most disproportionately are enokidake (×24.2 chance, n=48), kelp (×21.1 chance, n=71), barley (×19.0 chance, n=98), chinese cabbage (×16.8 chance, n=54), sake (×12.1 chance, n=154), sesame oil (×11.4 chance, n=392), squid (×10.6 chance, n=47) and shiitake (×10.4 chance, n=116). The score behind the ranking is positive pointwise mutual information — how many times more often a pair shares a recipe than random combination would predict — and n is the number of recipes containing both ingredients. Nothing is published below a floor of 20 recipes per pair and 100 per ingredient. Co-occurrence records what recipe authors actually do — it conflates deliciousness with popularity and tradition — so read it as ethnography, not judgment.

The measured partners — PPMI · ×chance · n

  1. enokidake — PPMI 3.19 · ×24.2 chance · n=48 (enokidake itself appears in 106 recipes)
  2. kelp — PPMI 3.05 · ×21.1 chance · n=71 (kelp itself appears in 180 recipes)
  3. barley — PPMI 2.94 · ×19.0 chance · n=98 (barley itself appears in 276 recipes)
  4. chinese cabbage — PPMI 2.82 · ×16.8 chance · n=54 (chinese cabbage itself appears in 172 recipes)
  5. sake — PPMI 2.49 · ×12.1 chance · n=154 (sake itself appears in 681 recipes)
  6. sesame oil — PPMI 2.43 · ×11.4 chance · n=392 (sesame oil itself appears in 1,843 recipes)
  7. squid — PPMI 2.36 · ×10.6 chance · n=47 (squid itself appears in 238 recipes)
  8. shiitake — PPMI 2.34 · ×10.4 chance · n=116 (shiitake itself appears in 596 recipes)

Cook the pairing

Dishes in Palate's live corpus that use soybean with one of its measured partners:

How this was measured

soybean appears in 1,192 of the 63,696 recipes counted (5,296 parsed + 1,081 hand-curated + 57,319 from the dataset published with Ahn et al., 2011). PPMI = max(0, ln(count(A,B)·N / (count(A)·count(B)))), all counts over the same N; lift = e^PMI, which reads directly as "×chance." Staples (salt, sugar, flour, water) are excluded because sources under-record them, and pairs where one name contains the other are excluded as near-duplicates.

Read the numbers with their limits: the corpus is Western-skewed (90% of it is the largely North American and European recipe dataset published with Ahn et al.'s 2011 paper), so the absence of a pairing here proves nothing except who wrote the recipes down. PPMI favors the rare end even above the floor — a niche pair at n=22 can outrank a workhorse pair seen thousands of times, because common ingredients' huge counts divide the score down — so read PPMI and n together.

The full reference — 150 ingredients, top 8 partners each — lives at What actually goes with X.

Curious how these flavors fit your palate — take the short taste test →

Every number above is counted from recipes — pointwise mutual information over 63,696 of them — not chef interviews, model guesses, or reviews.