Google Implementing Artificial Intelligence Elements Into The Search Engine

Artificial IntelligenceWhen Amit Singhal opens his mouth, you better pay close attention to what he has to say. He is the head Google search ranking team which basically makes him one of the most powerful people in the technological world. Yesterday, Amit opened his mouth. Are you all ears?

In a conversation with Lance Ulanoff from Mashable, Amit has discussed about how Google perceived the future of search and what the company is doing to reach there. Now, If you lived on this roundish ball thing we call Earth for the last 15 years or so, you probably know that Google always plan big.

The Situation Until Recently

When a user was searching for something just until recently, the search engine was “simply” matching his search query words to the best pages online where these words appears. “Simply” matching is obviously everything else than simple- The search algorithm constitutes more than 200 different rankings signals that calculates the results in less than one second.

So why did i used the word “simply”? To emphasis that eventually the search engine doesn’t really comprehended what stands behind all these different letters attached together, but only correlates between them to bring the most suitable pages with those words inside of them and in what words (anchor text) other websites are linking to these pages.

For example, if you searched for “Chicago jobs”, the search engine didn’t really knew what is it Chicago and what are jobs. All it knew is that it suppose to result the best websites or pages with the words Chicago and jobs. But Google already began the process of “teaching” its algorithm to go beyond words into entities.

The Process Already Began

Google intends to make its search engine to truly grasp the meaning of the word and to convert it into entity, just like humans are doing. You probably heard about this kind of “thinking computer” as Artificial Intelligence (AI). AI search engine that will understand entities, can provide much more relevant and deeper results for the user.

Obviously, just to develop this sort of much more sophisticated technology requires a lot of resources, manpower and also a vision that only a few tech companies in the world have. And while Apple has Siri and Microsoft has Kinect, you can be sure that Google already began working on something of its own. As Amit Singhal stated:

“Google is building a huge, in-house understanding of what an entity is and a repository of what entities are in the world and what should you know about those entities.”

Apparently, Google already began implementing few AI elements into its algorithm. For example, here’s what appears at the bottom of the results page whenever searching for “Leonardo da Vinci”:

Google Leonardo da Vinci

This showing that the search engine understands that Leonardo da Vinci was an artist and it manages to attribute artworks to the query. We can expect that this kind of advanced search “understanding” would only increase and improve through time. Amit didn’t hide his own reflections about it: “You can process query or question much better, and you get a step closer to building the Star Trek computer.”

I have to admit that I am feeling a bit funny now after being somewhat excited from Bing’s Duane Forrester recent video where he hinted Bing will grant social signals greater role. But now, Google proves (again) it is the real deal when it comes to search.

Update (3.15.2012)- In a new interview, Amit Singhal has stated that Google is about to launch over the next few months one of its biggest search upgrades in the history of the search engine, integrating much more artificial intelligence into the search engine. You can read the full story here.

Advertisement