|
|
Researchers at the University of Geneva have developed a tool that uses artificial intelligence (AI) to recognise counterfeit wines. The scientists used gas chromatography, an analysis technique for separating and recognising the chemical components of a wine, to trace it back to its origin. They analysed 80 wines produced over twelve years by seven different wineries in Bordeaux to create a reliable signature for each wine.
According to The Guardian newspaper, Professor Alexandre Pouget from the University of Geneva said that the AI was able to recognise specific châteaux based on cluster connections. It is able to match each "chemical signature specific to each château, regardless of vintage". "It is the overall pattern of concentrations of many, many molecules that characterises a château. Every château is a symphony: it is not a single note that characterises it, but the whole melody," Pouget continues.
The AI used in the programme could therefore detect counterfeit wines by tracing the links found in the wine back to specific châteaux. While the programme was able to match the wines to the correct châteaux with an accuracy of 99 percent, it had difficulties distinguishing the vintages and only achieved an accuracy of 50 percent.
The research results were published in the journal Communications Chemistry.
(ru / drinksbusiness)