PDA

View Full Version : OT -Cuil search engine now auto generates articles


prunch
04-09-2010, 12:18 PM
Check out this incoherent babble on the broncos. Its trying to take content from various sites and automatically make a coherent article.

http://www.cuil.com/search?q=denver+broncos

broncocalijohn
04-09-2010, 12:23 PM
why are they talking about Shanny and Cutler? Old news.

Man-Goblin
04-09-2010, 12:45 PM
This is actually a pretty cool idea, but there doesn't seem to be any rhyme, reason or timeline for the information that is pulled. Not sure how they can do it without accrediting the source websites, but it's a good start and hopefully a work in progress.

snowspot66
04-09-2010, 01:26 PM
This is great news. Now the sports writers won't have to work so hard to deliver all of those in-depth and highly analytical articles.

Cool Breeze
04-09-2010, 04:36 PM
This is great news. Now the sports writers won't have to work so hard to deliver all of those in-depth and highly analytical articles.

It reads like some we've seen.ROFL!

Mr.Meanie
04-09-2010, 04:39 PM
I thought I had stumbled onto a Woody Paige article for a minute.

chickennob2
04-09-2010, 04:54 PM
This is a dumb idea. They're essentially trying to design software that can search for facts/news about a topic and combine it into an encyclopedia article. That is a task people are great at. Trying to write code that does things people are good at makes sense in a lot of situations. Generally, these are situations where there is a very large or unknowable number of possible values and where results are needed immediately. For example: tone semantics in product reviews or speech recognition. Areas where this doesn't make sense are generally problems where only a finite number of operations need to be done (search terms in this case) and people are able to (and do) easily accomplish these tasks. For example: Encyclopedia articles.

Automated Wikipedia is a dumb idea. Why code things to automate a task that people A. are better at, and B. readily do for free.

Dumb.

If this is your PhD thesis, fine. If this is your startup: fail.