In the ultimate “it’s just not cricket” news, Belgian researchers have used a machine learning method to make better tasting alcoholic and non-alcoholic beers.
Researchers from Belgium’s University of Leuven characterised more than 200 chemical properties from 250 Belgian beers across 22 different beer styles.
They linked these to descriptive sensory profiling data from a trained tasting panel of 16 people and data from more than 180,000 consumer reviews from the online beer review database RateBeer.
With this large dataset, the authors trained machine learning models to correlate and predict flavour and consumer appreciation from the beers’ chemical profile.
They tested the effectiveness of the model by using its predictions to modify an alcoholic and non-alcoholic commercial beer, which achieved higher overall appreciation among trained panellists in blind tastings.
Lead researcher, Professor Kevin Verstrepen, said the tool could help improve quality control and recipe development of beers, or potentially other food and beverages, to meet specific consumer demands more efficiently.
“Understanding and predicting whether consumers will enjoy new food flavours is a complex task that is influenced by numerous chemical compounds and external factors. This presents a challenge in deciphering the relationship between beer chemistry and consumer preferences. Typically, this is investigated through consumer trials, which can be limited and somewhat inefficient,” the authors said.
The results are still restricted to major commercial Belgian beer styles. A larger number of samples is also needed to optimise predictions and overcome limitations, including identifying style-specific effects and demographic information, such as age and culture.
