Microsoft offering prize for better spelling correction

speller challenge

Spelling alteration is a technique search engines use to help you get better search results. It fixes errors when you misspell something, it can show alternative spellings of words and it can find synonyms of words you are searching for to help get better results. 

It seems Bing feels that there is a lot of room for improvement in this area of their search engine as they are teaming up with Microsoft Research to offer a reward to the team that develops the best spelling alteration technique. They are calling it the Speller Challenge and goal is to "develop a spelling alteration system suitable for large-scale statistical data mining-based web search."

A common approach to spelling alteration is the noisy channel model, in which the received query (q) is treated as a noise-corrupted version of the target query (c). In this model, the spelling alteration system alters q into c and returns the latter's results. The techniques to best identify query/target pairs and best estimate these statistics are the active research problem that underlies this challenge. But that's just the foundation. Place the spelling alteration task in the context of web search, and you have another dimension to consider. The effectiveness of using a fixed lexicon is a known problem because it can lead the speller not only to miss "real word" errors but also misrecognize out-of-vocabulary tokens as errors.

If this challenge sounds like something up your alley Microsoft is offering a sample dataset for developers and researchers to use, a similar dataset will be used to test the effectiveness of the spelling alteration engine. There will be five winners with the first prize winner getting $10,000.

Report a problem with article
Previous Story

Portal 2 PS3 edition to get cross-platform play

Next Story

Liberty Media files lawsuit against Hotfile.com

26 Comments

Commenting is disabled on this article.

its a good way to find someone that they might be able to hire, also might be cheaper than to have a team of people working on the same thing

Brody McKee said,
^ All this talk of spell checking in IE, what about the Windows Platform as a whole???

They can't because the EU would start crying.

xTdub said,
They can't because the EU would start crying.
Not sure so ... is there any competing OS-wide spell checking software?

considering the usefulness of such an algorithm, and that they'd instantly (US software) patent it to prevent other applications (office) using it . I think 10k is a terribly low prize.

yakumo said,
considering the usefulness of such an algorithm, and that they'd instantly (US software) patent it to prevent other applications (office) using it . I think 10k is a terribly low prize.

You probably get instantly hired to lol

I looked at the contest page, and I couldn't even understand the Rules.

Possibly this is a much more difficult problem than I ever appreciated.

Microsoft definitely needs to incorporate spell checking into Internet Explorer. It's the only browser on the market that lacks spell checking capability.

They would probably have less spelling problems in Bing if the IE team implemented a spell checker in their browser....

Omen1393 said,
They would probably have less spelling problems in Bing if the IE team implemented a spell checker in their browser....

Fewer, not less.
Judging by many of the posts on Neowin, having the option of a spellchecker doesn't do a whole hell of a lot. As the saying goes, you can't fix stupid.

Omen1393 said,
They would probably have less spelling problems in Bing if the IE team implemented a spell checker in their browser....

Here you go, a spell checking solution for IE; speckie.com

Relativity_17 said,

Fewer, not less.
Judging by many of the posts on Neowin, having the option of a spellchecker doesn't do a whole hell of a lot. As the saying goes, you can't fix stupid.

stupidity*

Shadowzz said,

stupidity*

It isn't wrong within the context of the sentence, since I was referencing a specific colloquial phrase. The accuracy of the reference takes precedent over grammatical shortcomings of the content (otherwise, there wouldn't be any point). Sure, I didn't follow all the technical tenets of the English language, but I'm better at it than most people.

Relativity_17 said,

Fewer, not less.
Judging by many of the posts on Neowin, having the option of a spellchecker doesn't do a whole hell of a lot. As the saying goes, you can't fix stupid.

You understood the meaning did you not? So what if he used a mass adjective with a count noun. Highlighting it only serves to reveal your grammatical pedantry.

Relativity_17 said,

It isn't wrong within the context of the sentence, since I was referencing a specific colloquial phrase. The accuracy of the reference takes precedent over grammatical shortcomings of the content (otherwise, there wouldn't be any point). Sure, I didn't follow all the technical tenets of the English language, but I'm better at it than most people.

precedence*

Sorry, couldn't resist! :-P

Neobond said,

Jeez that took AGES to install??

It takes a while because it creates a Windows Restore Point, incase things go bad, which they won't, but I like to play it safe