Microsoft Uses Human Computing Game To Tune Bing
Al writes "Microsoft researchers have come up with a novel way to fine-tune the algorithms behind the company's new search engine, Bing: a game that harnesses human computing power to improve the results. Called Page Hunt, the game (which of course requires Silverlight to run) shows users a web page and asks them to figure out a search query that should produce the page within the first five results. The idea is to better understand user behavior and expectations and ultimately improve its search algorithms. Other human-computing projects have sought to digitize out-of-print text (reCAPTCHA) and image labeling (Google Image Labeler). Can Microsoft use a similar approach to gain the edge over its rival? Or does Google already have the edge with SearchWiki, which lets searchers re-rank its results?"
So they're anchoring the algorithm in real-world data? Truly groundbreaking...
After the success of Page Hunt, Microsoft is developing a sequel called File Reports. Players earn points by filling out real business forms and increasing productivity!
That'll put a spoke in their wheel.
I have a valid excuse to surf porn.
The game gets boring really quickly the first time you run out of "reasonable" search terms and just tack on some exact quote from the page. "His father dies during the travel" is probably not going to help them very much, but it *will* get you to a specific bio of Paul Gauguin.
If users have the ability to tailor search results, won't page rank "fixers" (aka spammers) have an easier time? Or am I missing something?
Rhymes that keep their secrets will unfold behind the clouds.There upon the rainbow is the answer to a neverending story
This must be the beta. I don't see mention of a monthly charge, yet.
...the game (which of course requires Silverlight to run) shows users a webpage and asks them to figure out a search query that should produce the page within the first 5 results.
Gee, that sounds SO much more fun than playing the Sims! Not. reCAPTCHA works only because the user wants to get to what's after it, and doesn't require another downloaded plugin or frequent interaction. Guys, learn one of the great rules in IT: Just because it can be done, doesn't mean it should. If you want to investigate user behavior, do what everybody else in the industry has done -- install malware onto the user's machines and track their habits. :\
#fuckbeta #iamslashdot #dicemustdie
shows users a webpage and asks them to figure out a search query that should produce the page within the first 5 results
How much am I being paid? I suppose it is recession after all..
I sure hope no one tells 4chan about this.
I will stick my karmic neck out and humor you.
Thanks.
Dedicated Cthulhu Cultist since 4523 BC.
In theory, even if the venture is successul, what you will get is a search engine that understands gamers well. Is that going to improve your market share?
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
In other words, Microsoft service does not have solution. Microsoft asks users to provide results for themselves...
All this made me realise is how terrible Bing's search is. I mean... some of the queries failed to return the correct site, and I was literally "spelling it out" (full name of the page complete with some of the exact sentences/phrases on it).
If anything, this just makes Bing seem like a lost cause - it made the 'game' seem unfair (the engine was failing, not me) and completely pointless
I've got better things to do with my time than to do MS' job for them while having to install Silverlight at the same time.
Though without Silverlight, it may have been fun to come up with search queries for innocent things that involve gay animal sex, clown shoes and old people's inability to control their bowels.
It's actually quite interesting that they're asking for search terms rather than simply labels. Search engines are the a form of machine learning, and a lot of ML research goes into improving them. So it's interesting to consider what Microsoft is asking, in the context of ML. For example, Google has a game where users play by tagging images. Obviously, they're using some sort of supervised classification algorithm under the hood. But with Bing they're not asking for 'tags,' which would imply a supervised classification system, but search queries which return the page. Now that suggests that Bing is actually built on a bayesian model, which is very different from Google's markov steady state (page rank) model.
This "game" is about as much fun to play as those "fun with subtraction" pages in the fourth grade arithmetic book. It's bewildering, pointless, and laborious. And as nearly as I can tell there are no prizes. It's too clever by half.
A straightforward feedback link asking whether an ordinary Bing search got you the results you wanted would surely be more effective. Better yet would be an option to submit failed Bing searches to a human being who would attempt to find the answer and email it to you.
"How to Do Nothing," kids activities, back in print!
Or does Google already have the edge with SearchWiki, which lets searchers re-rank its results.
Anybody who has used SearchWiki to re-rank Google results, hold you hand up. Up high. Keep 'em up. Anyone? I didn't think so.
That last thing I need is "real people" screwing up the tags on a site. I recently got a new PC for home theater, and installed absolutely as little as possible on it (not even firefox - heresy, I know). I used the default search the first couple of times - forgetting that it wasn't google - and was amazed at how poorly the results came back. Even specific text known to be on the page (down to filenames I was trying to find for installing necessary codecs) wouldn't bring up the pages I needed. I can only assume that with (primarily) non-technical people typing in search keywords for pages it will just get worse.
You might say that a decade and a half of old search engine technology has trained me to make computer-based queries, but damnit it works, and I don't look forward to the unwashed masses breaking it.
Is it just my observation, or are there way too many stupid people in the world?