Analyzing Wine: Man vs. Machine
It appears that European students have gotten a little jump on Australia’s Commonwealth Scientific and Industrial Research Organisation CSIRO, who have also been working on a virtual nose to identify specifics in wine; region, varietal, age, flavors, and even price. About 50 students from 7 universities in Europe competed in the FOSS challenge, a competition to reprogram the FOSS winescan analyzer, normally used during winemaking, to provide details about any wine. The winning team comprised of Kim Houng Ngo and Martin Andersen from Aalborg University, Denmark, put the product up against a team of wine experts and came up with almost identical results regarding the region, quality, and price level. The experts came closer on price, but the students produced results much faster.
August 4th, 2006 at 8:51 am
[...] With the prices of wine futures soaring, and the large international market for rare bottles growing, fraudulent scammers and crooks are sure to take advantage of people by selling impostors and fakes. Traditionally, to combat fraudulent wine from being sold, professional tasters have been called in to make a determination on a bottle by tasting and comparing the wine for validity against his or her palate. Now, that process is closer to being automated by computer, as scientists from NEC’s System Technologies laboratory and Mie University, both in Japan, have developed a robot capable of comparing and identifying the unique characteristics that make up 30 different wines, with a larger field of recognition promised in the near future. Earlier in the year, students in Europe were essentially able to do the same thing, but the scientists were able to take it a step further by providing a comparison table for each bottle of wine analyzed. [...]