« October 2008 | Main | February 2009 »

January 14, 2009

AnswerFarm Technology from Ask.com

I recently blogged about advancements we’re making in semantic search technology here at Ask.com – that we’ve been testing our DADS, DAFS, and AnswerFarm technologies on NASCAR because it’s rich with structured data.

Now that we’ve entered NASCAR through partnerships, we’re going to apply our search technologies to this category even more deeply.  In addition to DADS technology, we think our AnswerFarm technology can be leveraged in an interesting way to help users learn more about this sport.

Suppose you are new to NASCAR (like I am).  There’s a lot of information about this sport on the web today and there are some very good content sources to get started.  But, we’ve been toying with the idea that learning about the sport may best be accomplished by “seeing it” through NASCAR’s millions of loyal fans.  In the Ask Q&A channel (click on “lots of answers” on our homepage to get there), you can see all of the different questions being asked about the sport across the web: What does NASCAR stand for? Why is NASCAR so popular? Are NASCAR drivers athletes? ...It’s like being a fly on the wall in a room full of lots of interesting chatter. 

The technology behind the Ask Q&A channel is called AnswerFarm technology. We built it by crawling and extracting question/answer pairs from across the web – more than 100 million question/answer pairs from several hundred thousand sources – and it is, no doubt, the most comprehensive and diverse repository of question/answer pairs in the world.

Here’s another example: search for NASCAR superstar Jeff Gordon. The Ask Q&A channel has more than 1,400 question/answer pairs collected from across the web… a massive Jeff Gordon “FAQ.” The Q&A channel is unedited, sometimes a little noisy… and therefore a lively and candid barometer of what people are discussing on any given topic.

Take a look at Jeff Gordon…
http://www.ask.com/ans?q=jeff+gordon&o=0&l=dir&qsrc=2872

Why Jeff Gordon is Jealous of Carl Edwards' Yaw 

…and Tony Stewart
http://www.ask.com/ans?q=Tony+Stewart&qsrc=2454&o=0&l=dir

Why do they call Tony Stewart "Smoke"?

…and Bobby Labonte
What do you think Bobby Labonte will do? (which we’ve answered… read today’s press release!)

The Ask Q&A channel covers a whole world of topics, and we’re building it into the largest repository of answered questions in the world.

Stay tuned for more search technology updates from us. My colleague Keith Hogan, VP of Search Development, will be speaking at the SMX West conference in a few weeks. He’ll be blogging about that, and will update you on other search technology initiatives we’re working on here at Ask. Stay tuned!

Tomasz Imielinski
EVP of Global Search and Answers

Posted by Ask.com Blog | Permalink | Comments (0) | TrackBack

January 08, 2009

Semantic Search Technology Advances from Ask.com

For millions of everyday queries, there is one, simple, direct answer the user is seeking, but the current generation of search relegates consumers to sifting through links and then searching through the Websites themselves to find the answer. Many times users have to repetitively rephrase the query before they finally get the answer they want. That’s because today's search engines are still very sensitive to the way queries are constructed, returning different answers for variations of the same query.

In October last year we introduced our proprietary DADS(SM) (Direct Answers from Databases), DAFS(SM) (Direct Answers from Search), and AnswerFarm(SM) technologies, which are breaking new ground in the areas of semantic, web text, and answer farm search technologies. Specifically, the increasing availability of structured data in the form of databases and XML feeds has fueled advances in our proprietary DADS technology.  With DADS, we no longer rely on text-matching simple keywords, but rather we parse users’ queries and then we form database queries which return answers from the structured data in real time.  Front and center. Our aspiration is to instantly deliver the correct answer no matter how you phrased your query.

Many of the categories consumers care about most are rich with structured data – meaning almost anything you’d want to know about that category exists in a database or XML feed somewhere. Extracting it for practical use is another matter.

TV Listings, which we blogged about in August, is a good example of how Ask.com is making exciting advances in semantic search – we’re giving consumers direct answers from structured data which was previously unavailable through any form of online search. Try searching Ask.com for “Football on TV this weekend” or “Movies on TV now.” No other search engine has been able to capture time-based data in this way, and we think it’s quite useful – much easier and more pleasing than scanning your TV’s onscreen guide. 

But America’s favorite pastime is sports, which are rich with real-time statistics, scores, records and trivia stored in the form of database tables. Users’ passion for access to athlete, team and performance data is still largely unmet by search engines due to the challenges of converting human queries to database queries against structured data.

We’re currently trialing our DADS, DAFS, and AnswerFarm technologies with NASCAR-related information, which we chose because the sport generates massive amounts of structured data. And queries about NASCAR continue to grow rapidly – it was one of the most-searched sports on Ask.com in 2008. Our goal is to give consumers answers instantly on the first results page to queries such as “who is the driver of car #60 in NASCAR?”

More importantly, we’re setting out to answer these types of queries regardless of how they’re phrased… something no search engine today can do.

Stay tuned for more semantic search technology advances from Ask.com.

Tomasz Imielinski
EVP of Global Search and Answers

Posted by Ask.com Blog | Permalink | Comments (0) | TrackBack

Opinions expressed here and in any corresponding comments are the personal opinions of the original authors, not of IAC Search & Media and may not have been reviewed in advance.

Blog Search from: Bloglines