Zemanta Launches Public API To Fuel Content Suggestion Engine Adoption
by Robin Wauters on December 9, 2008

Zemanta, the Slovenian startup that won the first Seedcamp competition (and subsequently funding from Union Square Ventures, Eden Ventures and The Accelerator Group), has just launched a public semantic API here at the Le Web conference in Paris, as well as a front side SDK.

Zemanta is basically a platform that aims to help online content producers find related content from across the web to enhance their publication(s). A blog, an article or a web page can be fed into its system, which then recognizes it and returns suggested images, smart links, keywords and relevant related stories from the web.

Based on the input, the engine recommends links to content from: Amazon, our own CrunchBase, Wikipedia, Freebase, MusicBrainz, Flickr, Daylife, and social networks.

With the launch of the public API, Zemanta is opening up its technology for blogging platforms, media companies, other startups and content database owners, in order to enable them to leverage their unstructured content and enhance it with relevant related content (see image above).

Usage of the API is free for up to 10.000 API calls per month, and for a subscription fee above that.

Here’s how they say it works:

We analyze your post through our proprietary natural language processing and semantic algorithms, and statistically compare its contextual framework to our preindexed database of content. We are using a combination of machine learning techniques and end-user input from our widget users, that enables us to train the engine and constantly improve the recommendations.

This is definitely technology that has lots of potential, and it will be interesting to see if Zemanta will be the one leading the advancements in this space.

Advertisement

Comments rss icon

  • Comments didn’t work previously, so responding late.

    Basically this is also the first API to disambiguate to Linkin Open Data entities, which means more than 3 million entities that a developer can then reason about further by querying dbPedia, Freebase, MusicBrainz or Semantic CrunchBase.

    We’re trying to take the feedback from early user community and improve the service through it.

    Andraz Tori, CTO at Zemanta

  • As the founding blogger at CrunchyCommerce.com, this looks like a huge power tool for my WordPress setup.

    I’m interested. I just might sign up.

  • We use Zemanta on our blog and it is fantastic. This is great news. Now we can integrate it with our main site.

    Well done guys.

  • Yawn. Boring.

    I think I will just stick with google which returns better results.

  • This API is dumb as a cow, it doesn’t weigh what is really important in a site. The only thing it does is a milkshake of keywords and then look for those in several public sites… I can do that as well without the ads! I was expecting something a bit more sophisticated.

  • We integrated Zemanta’s API on MyBlogLog while it was still in private beta several months ago. We use their system to analyze content we are pulling in from blogs (including blo.gs) and life stream data (with some help from Gnip) to categorize for our Topics system. The results have been fantastic.

    I highly recommend that everyone check out what Zemanta is doing — really cool/important stuff. And a great group of guys to boot.

  • We use this over on MyTriniPhone.com and it works wonderfully!!

  • Previous version of zemanta itself, they are doing great. This is a great step from zemanta, suggesion engine .. nice move..

  • We’ve been using the Zemanta API for our private beta for a few months now and it’s great.

    Moreover, you can see great progress and improvement each month so I’m sure this is just the tip of the iceberg for what the Zemanta API and app will help you do.

  • Zemanta is great. I use it all the time

  • Zemanta is very good met them at Seedcamp but why is this story being covered on TC.com and not on TCUK. Everytime there is a good story about UK/Europe TC.com covers it and the rest of the dull stuff is left for Butcher to cover.

    Make TCUK TC Europe and focus the contnet where it originates.

  • This is pretty cool. I’ve played with it a bit and get OK results.

    One note of caution – read the terms of service. Zemanta *owns* the content you send to them. Be careful you don’t send anything you don’t own or that you don’t want public. The way it’s written the content owners will end up coming after you.

  • This is a bit of misunderstanding. To make it clear: Zemanta does not own the content sent through the API, but it does retain some rights to reuse of the data (to improve the service, etc.). Please contact us at support@zemanta.com to clear up any issues. Maybe we should reword the TOS if it is not clear enough.

    Andraz Tori, Zemanta

Leave Comment

Commenting Options

Enter your personal information to the left, or sign in with your Facebook account by clicking the button below.

Alternatively, you can create an avatar that will appear whenever you leave a comment on a Gravatar-enabled blog.

Trackback URL
bugbugbugbug
Techcrunch on Facebook