The research comes just a few months after Google Search’s Big Data was dismissed as a predictive tool, following reports that Google Flu Trends was well off the mark when it came to guessing cases of influenza.
But that small hiccup hasn’t stopped academics from the University of Warwick in the UK from attempting to do the same with stock markets. In a newly published paper, the researchers explain that by relying on the emergence of phrases on Wikipedia that were associated with previous stock market crashes, and combining this with Google search terms, they can predict when financial markets are likely to take a tumble.
“First, we take a large online corpus, Wikipedia, and use a well-known technique from computational linguistics to identify lists of words constituting semantic topics within this corpus,” the researchers explain.
The researchers then enlisted the help of Mechanical Turk users to give names to each of these topics. They followed this by examining the incidence of these topics showing up in Google searches from 2004 all the way up to 2012, to see if they could spot any relationship between the volume of these searches and stock market movements.
Their findings show that “for complex events such as financial market movements, valuable information may be contained in search engine data for keywords with less-obvious semantic connections”. Basically, the researchers believe they’ve spotted a correlation between the things people search for on Google before they make a buy-sell decision. They also believe this technique can be applied to many other situations. “We suggest that extensions of these analyses could offer insight into large-scale information flow before a range of real-world events”, the study notes.
Such a technique could obviously be beneficial for those who like to play the stock markets, but before they get too excited, it should be noted that the research focused on Google Trend’s aggregations of search terms, rather than people’s actual queries. The study also notes that the data would only be marginally more reliable than a regular “buy-and-hold” strategy, for example.
In other words, there’s probably a long way to go before investors start using Google to shape their stock market strategies.
photo credit: Perpetualtourist2000 via photopin cc
About Mike Wheatley
Mike loves to talk about Big Data, the Internet of Things, Hacktivists and hacking, but he also hates Google and can never resist having a quick dig at them should the opportunity arise 🙂 Got a REAL news story or tip? Email [email protected]