Search engines rely on algorithms to determine relevant results to a user’s query. Algorithms fulfill all sorts of different functions related to data display and data mining, but up until now the most widely practiced way has been through Google and search engines.
Today, we are looking at the possibility to algorithms that can accurately predict our future. Don’t believe it? Check out these REAL examples of algorithms that can predict future events accurately.
Brad Brodigan, SVP and General Manager of Attensity, has been monitoring your conversations. "IdolStats.com provides a fun example of the power behind Attensity360's advanced social media monitoring technology and the predictive results that we can deliver to our clients."
IdolStats measured millions of social media interactions (Twitter, blogs, comments etc.) about American Idol in order to gauge American sentiment toward contestants. The service, called Attensity360, automatically collects and crawls chatter happening on blogs, message boards, forums, micro-blogs, online news media and social networks. IdolStats.com then ranks contestants based on negative or positive sentiment and their share of the total American Idol conversation.
Selecting the American Idol cast off is just a small example of what Attensity Group's acquisition of Biz360, enables them to offer to big business. The technology enables businesses to effectively listen, analyze, relate, and act on information, both inside and outside of the firewall.
AOL’s Story Factory
The Wall Street Journal caught a comment from the CEO of AOL, Tim Armstrong, that you might find interesting. Word is that the ISP turned news source will be using “a new digital-newsroom system that uses a series of algorithms to predict the types of stories, videos and photos that will be most popular with consumers and marketers.”
A sample graph outlines the history of home values. Will graphs like this help determine future costs?
The obvious way this is done relies on collecting search data and ranking it in order of importance. Will this predict and create stories? Who knows… how long until digital news filters like this are reporting ridiculous YouTube videos at the front page? The last thing I need is Lol Catz on the front page of my news source because the algorithm can’t tell the difference between viral and newsworthy.
Using Algorithm’s to Predict Malicious Attacks
By comparing the history of attacks you system has suffered against the behavior of other users, University of Irvine scientists hope to develop an algorithm that effectively blacklists sites that have a higher probability of attacking users with malicious viruses. The process is comparative to the way Amazon will recommend books you might like based on the way other users react to the items you are currently searching for.
At this point, the team claims a 70% success rate. While this is a good sign, the team has been unable to rule out the human factor of choice (where to attack from) and the way attackers’ patterns work. Basically the system is not air tight, but as a kind of safety guidebook, the scientists may be on to something…
Predicting the Next Music Hit
According to Popular Science, algorithms are being developed in Tel Aviv that allow researchers to predetermine songs that will be major musical sensations with 30-50% accuracy. The algorithm analyzes data pulled from Gnutella and crawls it to determine things like Soulja Boy rising to the top of the charts a full two months before they occur.
Algorithms and our Future
While this is not groundbreaking, it does make for some interesting conclusions. Looking at any particular stat correlated to human behavior, it is possible to analyze trends and make educated guesses about some of the elements of our future. Can we arrange an algorithm to accurately guess who will win the World Series? Probably not yet, but we can make for some very accurate guesses in many other areas.