You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  Current Article

Taking Text And Sentiment Analytics To The Masses

By   /  November 20, 2012  /  No Comments

Text and sentiment analytics for the masses. That could be a tagline for Semantria, which lets users put the technology to work in a pay-as-you-go cloud model. Not only that, but it lets customers deploy a plug-in to run analytics of unstructured content, extracting entities, themes, sentiment, categories, summaries, facets, and relationships, in one of the world’s most common user environments: Microsoft Excel.

More is on the way, too. This December should see the unveiling of a partnership with an as-yet-unnamed vendor to expand the applications with which its platform is compatible. That partner already offers data integrations with 300 applications; when Semantria becomes the 301st, users will be able to universally and bi-directionally talk to the hundreds of other applications without having to do any integration work on their own.

The Excel plug-in, says Semantria founder and CEO Oleg Rogynskyy, has found a home with a lot of organizations and individuals running small surveys or other activities where they need to make sense out of unstructured text but that don’t want to take on programming or integrating chores to have their content processed and ready for analysis. Users get an API key good for 10,000 free documents that they insert into the downloadable plug-in, and they’re on their way – no matter what they want to do. As an example, marketing director Rami Nuseir says he uses it to analyze movie tweets when he’s bored.

In fact, one blogger recently did just that, to get some deeper insight into the new James Bond movie, Skyfall, with an analysis of 3,000 tweets. Another use case: a university in Canada has 40,000 transcripts of people’s nightmares as part of a research project that’s trying to understand if there are common entities or themes appearing in nightmares among different demographics that correlate in time with real-life events.

“This kind of use case was not available before there was an out-of-the-box, easy-to-use-analytics tool that is cheap,” says Rogynskyy. The Excel plug-in actually was originally built as demonstration technology to give people a way to try out the main Semantria API solution, a RESTful API to process text in real-time that can handle any amount of traffic.  Users that want direct access to the API tend to be resellers like social media monitoring companies or survey providers, who build a solution using the Semantria technology for their own customers.

The pricing model – unique in the industry, he says – follows the pre-paid cell phone idea. A minimum buy-in of $1,000 lets users process 100,000 documents (one cycle of processing, in the vendor’s terms). Users can reload more credits as they need them, and if they don’t use all the credits in one project, they stay in their account without an expiration date. That helps customers plan expenses, he says, and in particular helps agency-customers know exactly how much they spend on text analytics for one project vs. another one, to parse exact costs down to the individual customer.

Semantria’s technology is in fact Lexalytics’ Salience Analytics engine, and it was the first API product to announce in November sentiment analysis support in five languages (adding Spanish, Portuguese, and French to English and German) based on its deploying the 5.1 version of Lexalytics’ technology for its offering. Rogynskyy also is the head of business development and marketing at that company, which is a Semantria investor. To that end, it shares all the capabilities of that technology – including its Wikipedia ontology for categorization, classic rules-based categorizer, and collections mode to analayze most frequently-used keywords, they keywords related to them and sentiment. Only it does it on the Amazon cloud, with true multi-tenancy and almost-infinite scalability – plus it takes advantage of the global view of sentiment and categorization provided by the hundreds of accounts for which it is processing all kinds of data, to auto-curate and improve those capabilities based on that feedback.

Also in the pipeline for Semantria is a crawler, so that “at some pt our customers can give us a set of URLs and we provide them with a whole bunch of content, sentiment, and so on about them,” Rogynskyy says. And it’s also currently worktin with Lexalytics on adding Chinese language support for sometime early in the New Year.



About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Is Data Governance Solely About Controls on Data?

Read More →