Time to Give up the Day Job: Technology is Much Better than Humans at Predicting the Value of Wines


Posted on Mon 21st Sep 2015 at 12:32




As anyone who has even looked into the possibility of collecting wines knows, predicting their future value is something of a dark art.


Or at least, was, until the boffins at UCL came up with a clever new way of making the predictions through machine learning. What's great news for collectors, and somewhat dismal news for those who make a living from making predictions the old fashioned way, is that the machines are 98% better than humans at gauging fine wines' future value. The reason the machines are so good is that they use something called machine learning, which allows them to identify which elements of the data are the key predictive factors for changes in price. Speaking to Harpers magazine, Professor John Shawe-Taylor, co-director of the UCL Centre for Computational Statistics and Machine Learning, said: “Machine learning involves developing algorithms that automatically learn from new data without human intervention. We’ve created intelligent software that searches the data for useful information which is then extracted and used, in this case for predicting the values of wines.” The team continues to work on their software, and hopes our industry will develop greater confidence in machine learning methods as time goes on. Although it may be a while before machine learning is considered a mainstream tool for making investment decisions, it seems like it could be a legitimately useful way of enabling investors to make informed decisions about their wine portfolios. For early adopters, the Invinio wine asset management consultancy, founded by another UCL boffin, is collaborating on the machine learning research. What do you think? Would you trust a machine over your own palette? We'd love to hear your views here.


Excellence as standard....that’s our Marque of Distinction


Our Opening Hours

Monday: 9:00am - 17:00pm
Tuesday: 9:00am - 17:00pm
Wednesday: 9:00am - 17:00pm
Thursday: 9:00am - 17:00pm
Friday: 9:00am - 17:00pm

Copyright © 2005 – 2024 Signature® , all rights reserved. Signature and the Signature logo are registered trademarks of Signature.