Google April 16 improved the efficiency of its search engine by refining Google Suggest and spell correction and adding auto-correction for different languages.
Google Suggest offers users queries in a pull-down menu when users begin typing in searches into the search box on Google.com. The search engine began offering localized searches by country in March 2009, surfacing relevant search queries for different regions.
Suggest later began showing suggestions relating to the current results page, personalized and navigational suggestions, and ad links. Google last December also added universal search results into Suggest.
Google has narrowed Suggest even further by tailoring suggested queries to a user’s home city.
In Google’s example, a user based in San Francisco who searches for “Bart” will see the suggestion for Bay Area Rapid Transit. For Chicagoans, a search for “bull” will surface suggestions for the Chicago Bulls.
Google engineer Pandu Nayak explained this added localization thusly: “Just as people in the U.K. often look for different things than people in U.S., we’ve found that people in Seattle tend to look for different things than people in Dallas.”
Google also said it is getting better at suggesting corrections for searches users mistype (users see this as the “Did you mean” option when they make an error).
Specifically, Google is providing some relief for peoples’ names, which users sometimes butcher when they type them.
Now when users search on names, Google will recognize descriptors, such as occupation, education or other contextual affiliations to help users find information on the right person.
“Users often include other terms such as ‘composer’ or ‘lawyer sparta wisconsin’ in their search query, which provides valuable context to help us narrow the range of possibilities for the spelling correction,” explained Google’s Nayak. “We use these additional descriptive words to offer you better suggestions.”
And this doesn’t just cover searches for celebrities or politicians. This will work whether users are searching for Joe Plumber or Barack Obama. Google provides examples such as “yuri lehner stanford” and “simon tung machine learning.”
Improved spelling correction is available in Google’s English spelling system in the United States, but the company will roll out the enhancement to other parts of the world and other languages in the coming months.
Finally, Google added auto-correction to its spelling system in 31 languages across 180 domains. Users who search for “aiprt” will be whisked straight to the results for the corrected search, rather than be offered a link on the results page asking them if they meant to search for “airport.”
By narrowing the error margin, Google will help users find what they’re looking for faster. Ideally, users will find what they need and move on to do the next search.
The faster Google serves searches, the more users will search. The more users search, the more ads they will see. Putting more ads in front of users means more money for Google’s AdWords program.
Google’s emphasis on speed has helped it rake in $22 billion a year in advertising and helped it post a 37 percent profit hike in the first quarter of 2010.