I have a BlackBerry Curve with Google installed. One of the cool aspects of this search program is that it allows me to search by voice. I speak into the phone while holding the send button, and Google chews on what I said and, hopefully, gives me the right results. The first time I tried it, I searched for "Blaine Gabbert", the current quarterback for the Missouri Tigers. Sure enough, I got the right results back.
Recently, feeling a bit saucy, I thought I'd try and trip up the program by searching for my wife. My wife is a pharmacist and has a doctor of pharmacy degree, aka a "pharm d." So I said her name and "pharm d" into the speaker and thought "I've got you now, smart guy."
Phil = FAIL. Google = WIN. It got the words correct. How cool is that?
Want to know how Google has solved problems like this? By employing lots of really, really smart people, by making mistakes and learning from them, and by not giving into the fact that it operates in a very competitive environment, despite having a 65% market share. Here's the story of Google's algorithm (via Craig Newmark). Here's an excerpt.
Take, for instance, the way Google’s engine learns which words are
synonyms. “We discovered a nifty thing very early on,” Singhal says.
“People change words in their queries. So someone would say, ‘pictures
of dogs,’ and then they’d say, ‘pictures of puppies.’ So that told us
that maybe ‘dogs’ and ‘puppies’ were interchangeable. We also learned
that when you boil water, it’s hot water. We were relearning semantics
from humans, and that was a great advance.”
But there were obstacles. Google’s synonym system understood that a
dog was similar to a puppy and that boiling water was hot. But it also
concluded that a hot dog was the same as a boiling puppy. The problem
was fixed in late 2002 by a breakthrough based on philosopher Ludwig
Wittgenstein’s theories
about how words are defined by context. As Google crawled and archived
billions of documents and Web pages, it analyzed what words were close
to each other. “Hot dog” would be found in searches that also contained
“bread” and “mustard” and “baseball games” — not poached pooches. That
helped the algorithm understand what “hot dog” — and millions of other
terms — meant. “Today, if you type ‘Gandhi bio,’ we know that bio means
biography,” Singhal says. “And if you type ‘bio warfare,’ it means
biological.”