In the Bordeaux region of southwest France, vineyards estates transform grapes into bold red wine blends. Some bottles sell for thousands of dollars each. Prestigious chateaus boast about the soil, microclimate and traditional methods that make their wine superior, an inscrutable mix known as terroir. A computational neuroscientist is trying to apply chemical precision to this je ne sais quoi, which is one of the terms that the wine industry likes to keep a bit mysterious. In a study published in the journal Communications Chemistry, the neuroscientist and his colleagues described a computer model that could pinpoint which Bordeaux estate produced a wine based only on its chemical makeup. Although wine connoisseurs often claim to be able to distinguish among wines from top estates, they rarely do blind taste tests. The neuroscientist grew up in Paris in a family that drank only Bordeaux and believes that machine learning methods could be useful for the wine industry. He teamed up with a scientist from the Institute of Vine and Wine Science in Bordeaux to create a database of 80 wines from seven chateaus. The researchers trained an algorithm to seek common patterns in the chemical fingerprints of the wines. Independent researchers said that the study was part of a wave of recent research using machine learning to decipher terroir. Another application of these models is rooting out fraud, which is fairly common among expensive wines. Experts believe that the approach would work for any wine region if the model has been trained on a large array of wines from different producers and vintages.
