AI Helps Optimize Beer Taste and Composition, But Brewers' Skill Still Vital
-
Researchers used AI and machine learning models to analyze the chemical makeup of 250 Belgian beers and predict taste/appreciation based on composition.
-
A tasting panel sampled and rated the 250 beers on 50 attributes over 3 years to provide data for the models.
-
Online beer reviews were analyzed to supplement the tasting panel data.
-
The models were used to tweak an existing beer's composition and improved taste ratings, especially for non-alcoholic beers.
-
While AI can optimize beer composition, brewers' skill and expertise remains vital in beer production.