Last week we discussed some of the specialized and niche search engines that have emerged to compete with Google. This week an article in Wired explains how Google is continually refining its search algorithms to provide the best results of a generalized search engine. One interesting method Google uses to refine their algorithms is to learn from how people word and rephrase their searches, allowing the algorithm to learn synonyms and to put words and phrases into context. In an example from the article:
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."
When Google engineers want to test a potential improvement of their search algorithm, they enlist everyday Google users in experimental groups to try it out. According to Wired, you're probably participating in some sort of experimental improvement nearly every time you run a search:
But Google also has a larger army of testers - its billions of users, virtually all of whom are unwittingly participating in its constant quality experiments. Every time engineers want to test a tweak, they run the new algorithm on a tiny percentage of random users, letting the rest of the site's searchers serve as a massive control group. There are so many changes to measure that Google has discarded the traditional scientific nostrum that only one experiment should be conducted at a time. "On most Google queries, you're actually in multiple control or experimental groups simultaneously," says search quality engineer Patrick Riley. Then he corrects himself. "Essentially," he says, "all the queries are involved in some test." In other words, just about every time you search on Google, you're a lab rat.
Using the information collected during its tests, Google makes frequent alterations to its platforms as well as major upgrades every few years (most recently with Google Caffeine).
Google is also staying competitive with other specialized search products. In addition to Google Squared for categorized search results, Google recently incorporated a real time search that canvasses recently published tweets and blog entries similar to Bing's Twitter search.
For more insight into Google's ongoing improvements, read the full article from Wired here.