Bernadette Tansey
Let’s say you’re a tech enterpreneur whose groundbreaking gizmo obtained rave reviews in influential publications, spurring months of beneficial gross sales. but later, gross sales slide downward, and also you don’t be aware of why.
turns out, a rumor is circulating in dialogue boards that your product has a excessive failure rate after six months of use—a improper affect or a deliberate lie. How would you become aware of those inaccurate comments as quickly as they give the impression of being, so that you would be able to publicly refute them?
It’s the kind of problem that Diffbot, a small Palo Alto, CA-based artificial intelligence company, got down to help companies meet. Diffbot introduced lately that it has created a brand new search instrument, Discussions API, that digs for product mentions in remark threads, group forums and online critiques—the hidden crannies of the “deep internet” it says Google doesn’t absolutely explore.
“Engaged commenters may be misrepresenting a brand,” Diffbot product executive John Davi says. “Now that’s easy to search out.”
whereas conventional search engines like google mechanically pull up lists of published articles about merchandise, Diffbot indexes the net conversation that follows below those stories. the amount of this person-generated commentary may be as much as four hundred occasions higher than the “surface internet” of mainstream media, organizational websites, and different content easily available with conventional engines like google, Diffbot says.
while firms regularly search Twitter feeds as part of their media monitoring routines, Davi says one of the most most influential client backchat may just take place in other forums where writers aren’t limited to 140-character messages. and those other comment sections don’t have their very own constructed-in search capabilities, as Twitter does. Diffbot is now making those boards searchable.
while a Google keyword search would possibly flip up a couple of individual feedback discovered on the internet, Diffbot scans the discussion sections throughout a couple of web sites and returns the implications to the customer in a database format that summarizes many comments, Davi says. through using synthetic intelligence and robotic expertise, Diffbot extracts key important points from each and every remark, together with the author, writer url, the website the place the comment appeared, and the character of the opinion expressed.
The database can determine sure tendencies in public opinion in addition to destructive misconceptions, Davi says. Product makers acquire the power to reach out to customers whose favorable comments could be utilized in firm advertising campaigns, he says.
dialogue API searches opinion forums including fb comments, Disqus, WordPress, Blogger, and Reddit.
“We if truth be told predict this to be a gorgeous sizzling commodity,” Davi says. the theory to look comment sections arose each from Diffbot’s body of workers and from its present buyers, he says.
Diffbot’s new Discussions instrument builds on different innovations the company has made to extend the searchable territory on the internet. founded in 2009, Diffbot first attacked a blind spot in the best way a robotic mind “reads” a website. unlike a human being, a traditional search engine can’t distinguish neatly between the totally different sections of a webpage, similar to a story headline, the author’s byline, an image, and the body of the story.
the use of desktop studying and human trainers, Diffbot taught its robot brain to recognize these layout components in plenty of different webpage types, reminiscent of a entrance page or dwelling page, an article, an e-commerce website online showing product information, and pages containing images or videos.
buyers comparable to Instapaper use this Diffbot structure-studying operate to reformat internet content for use on mobile devices. Diffbot routinely reshuffles the positions of headlines, text blocks, and other parts to suit the totally different dimensions of a smartphone or pill screen.
however Diffbot also uses the format-reading function to search out data such as product prices—its robotic brain now knows which page areas to seek for that knowledge. clients like Pinterest may use the price data Diffbot extracts from a couple of places to lend a hand its customers with comparability buying. Product wholesalers use Diffbot’s aggregate reports of present price knowledge to verify shops aren’t violating agreements to cost buyers the recommended manufacturer’s retail price, Davi says.
“We’ve constructed a relatively tidy business on that,” Davi says of the web page-studying capabilities that had been Diffbot’s first business services. He declined to divulge company revenue numbers. buyers of the 12-employee company embody Adobe, CBS Interactive, Cisco, eBay, Salesforce, and Samsung.
Diffbot has raised a total of $ 2 million because it used to be based, and doesn’t want to raise capital for some time, Davi says.
“We’ve had damage-even months, however then we keep hiring individuals,” he says.
Xconomy
(160)
Related
Tweet
<> Embed
@ Email
Report
Diffbot Turns online comments Into Market Intelligence Databases
Bernadette Tansey
Let’s say you’re a tech enterpreneur whose groundbreaking gizmo obtained rave reviews in influential publications, spurring months of beneficial gross sales. but later, gross sales slide downward, and also you don’t be aware of why.
turns out, a rumor is circulating in dialogue boards that your product has a excessive failure rate after six months of use—a improper affect or a deliberate lie. How would you become aware of those inaccurate comments as quickly as they give the impression of being, so that you would be able to publicly refute them?
It’s the kind of problem that Diffbot, a small Palo Alto, CA-based artificial intelligence company, got down to help companies meet. Diffbot introduced lately that it has created a brand new search instrument, Discussions API, that digs for product mentions in remark threads, group forums and online critiques—the hidden crannies of the “deep internet” it says Google doesn’t absolutely explore.
“Engaged commenters may be misrepresenting a brand,” Diffbot product executive John Davi says. “Now that’s easy to search out.”
whereas conventional search engines like google mechanically pull up lists of published articles about merchandise, Diffbot indexes the net conversation that follows below those stories. the amount of this person-generated commentary may be as much as four hundred occasions higher than the “surface internet” of mainstream media, organizational websites, and different content easily available with conventional engines like google, Diffbot says.
while firms regularly search Twitter feeds as part of their media monitoring routines, Davi says one of the most most influential client backchat may just take place in other forums where writers aren’t limited to 140-character messages. and those other comment sections don’t have their very own constructed-in search capabilities, as Twitter does. Diffbot is now making those boards searchable.
while a Google keyword search would possibly flip up a couple of individual feedback discovered on the internet, Diffbot scans the discussion sections throughout a couple of web sites and returns the implications to the customer in a database format that summarizes many comments, Davi says. through using synthetic intelligence and robotic expertise, Diffbot extracts key important points from each and every remark, together with the author, writer url, the website the place the comment appeared, and the character of the opinion expressed.
The database can determine sure tendencies in public opinion in addition to destructive misconceptions, Davi says. Product makers acquire the power to reach out to customers whose favorable comments could be utilized in firm advertising campaigns, he says.
dialogue API searches opinion forums including fb comments, Disqus, WordPress, Blogger, and Reddit.
“We if truth be told predict this to be a gorgeous sizzling commodity,” Davi says. the theory to look comment sections arose each from Diffbot’s body of workers and from its present buyers, he says.
Diffbot’s new Discussions instrument builds on different innovations the company has made to extend the searchable territory on the internet. founded in 2009, Diffbot first attacked a blind spot in the best way a robotic mind “reads” a website. unlike a human being, a traditional search engine can’t distinguish neatly between the totally different sections of a webpage, similar to a story headline, the author’s byline, an image, and the body of the story.
the use of desktop studying and human trainers, Diffbot taught its robot brain to recognize these layout components in plenty of different webpage types, reminiscent of a entrance page or dwelling page, an article, an e-commerce website online showing product information, and pages containing images or videos.
buyers comparable to Instapaper use this Diffbot structure-studying operate to reformat internet content for use on mobile devices. Diffbot routinely reshuffles the positions of headlines, text blocks, and other parts to suit the totally different dimensions of a smartphone or pill screen.
however Diffbot also uses the format-reading function to search out data such as product prices—its robotic brain now knows which page areas to seek for that knowledge. clients like Pinterest may use the price data Diffbot extracts from a couple of places to lend a hand its customers with comparability buying. Product wholesalers use Diffbot’s aggregate reports of present price knowledge to verify shops aren’t violating agreements to cost buyers the recommended manufacturer’s retail price, Davi says.
“We’ve constructed a relatively tidy business on that,” Davi says of the web page-studying capabilities that had been Diffbot’s first business services. He declined to divulge company revenue numbers. buyers of the 12-employee company embody Adobe, CBS Interactive, Cisco, eBay, Salesforce, and Samsung.
Diffbot has raised a total of $ 2 million because it used to be based, and doesn’t want to raise capital for some time, Davi says.
“We’ve had damage-even months, however then we keep hiring individuals,” he says.
(160)
Related
Pinned onto Intelligence
Embed Pin on Your Blog
Email This Pin
Report This Pin