In the summer of 2008, at an artificial intelligence confab deep in Silicon Valley, Yahoo! senior research scientist Deepak Agarwal revealed that the web giant was using automated algorithms to select news stories on its famous front page. These algorithms, he said, had boosted click-through-rates by 25 to 30 per cent, driving …
I! Love! The! Exclamation! Marks! In! Yahoo! Headlines!
Put! It! Down! To! Nostalgia! For! Bang! Addrtesses!
Not the 9 O Clock News
Most of Yahoo's news stories seem to be borrowed from the pages of the Metro (the free daily rag delivered to London's tube users), and about a day or two behind other news sources. Not really impressed with their news coverage.
Give it a second thought would ya
First of all, these are the kind of stories that provide click-through rate on Joe Average cittizen.
Second, quite clearly you are classed in that bucket. Either the system does not have enough information on you, or it the information it has collected shows that you are likely to click-through on Metro material. Recognise yourself? : http://www.thedailymash.co.uk/news/society/guardian-readers-finally-get-excuse-to-buy-a-tabloid-201107114058/
In any case, it further confirms that you make money buy showing the news "we would like to see" instead of showing the news.
Yahoo news "coverage"?
They don't so much cover the news as scrape it.
...that's because your personal profile indicates that you get all your news from two-day old papers you find on the Tube.
skipped the entire article so I could write this...
Who the hell still uses Yahoo?
Those of us who resist the beast...
I presume you're not talking about the Seattle beast
If you are, you're not doing a good job...
Their homepage click prediction algorithm isn't that complicated in my case
I always go straight into Classic Mail and ignore all the bollocks.
"But at the same time, it predicts clicks for individual users or segments of users, leaning on information such as their sex and their age (which users supply when they sign up for a Yahoo! account"
You mean there are people who don't fill those fields up with rubbish?
As long as people with similar tastes choose similar rubbish to fill in the fields the algorithm doesn't really care whether it's rubbish or not. That's why it's smarter than humans
"I always go straight into Classic Mail and ignore all the bollocks."
Damn, dude; you totally missed out on a great Sex Pistols reference, there.
Mine's the black one with the big red "Circle A" spray-painted on the back, thanks.
Those of us who are still resisting...
Resistance is futile!
I don't see what the fuss is about.
You select from a pool of stories, and optimise the most likely to be clicked to the top, based on readership.
Anything like "man gets hit in groin by footbal" goes to the top.
And something like "housefire kills family of six" goes to the bottom.
No-one's done this bit yet?
"...it predicts clicks [for users based on] what browser they use."
I can do that:
Opera users: ( well, we're to smart to fall for that ; P )
FireFox users: prefer anything with "Open Source" in it,
Safari users: go for anything with a picture of a partially-eaten fruit in it,
Chrome users: click on anything that asks for your personal information,
and [all together now]
IE users: Click On ALL The Malware Links!
Human-driven click prediction
"IE users: Click On ALL The Malware Links!"
Especially if they're rapidly flashing, announcing VIRUS DETECTED! CLICK HERE FOR A FREE ANALYSIS!
IE users: Click On ALL The Malware Links!
Double malware points for users infected with the Yahoo toolber
So what they are saying is that you get to read what the people before you read, divided into sets depending on what you are reading now? Is that it?
And the editors get asked for more Jo-Lo stories, or more pictures of sheep up trees, because that is what sells?
I just reclaimed my yahoo name, for grins, and found a cluttered portal, not a search engine. There must be some incompatibility with something I set in 1998.
You don't have to tell them your age or gender so their data is not reliable and therefore their results are not good.
Their groups are OK though.