Attensity Introduces Semantic Annotation to Interpret Unstructured Data
New SDK enables companies to quickly parse unstructured texts and better understand their customers' needs.
Posted Feb 26, 2015
Page 1

To help organizations make the most of their unstructured data, Attensity, a natural language processing company dealing in text analytics, today released its Semantic Annotation solution in combination with its software development kit (SDK).

The system collects the customer data that is generated through hard-to-track documents including social media posts, emails, and surveys, and runs them through a parsing engine. "The parsing engine parses at least 100 times faster than a human can, even if that person is an industry expert," Matt LeGare, product manager at Attensity, says. "A tweet takes the engine a thousandth of a second—or a tenth—depending on the content, whereas a human reading that same tweet would take thirty seconds." Since data scientists don't have the time to monitor every single instance in which their company is mentioned, the technology saves time by automating an otherwise impossible process for them. To equal the amount of work done automatically using this solution, LeGare says, firms would have to expend thousands of hours a week monitoring social media sites.

Building on advancements it made with the recently updated Attensity Q, the release will widen the reach of the company's offerings. LeGare says that, unlike social tracking products such as Salesforce.com's Radian6, which relies on a keyword approach, Attensity's product is unique in that it looks at all of the linguistic cues inside of a document and uses those to determine sentiment, interest, and categorization. "We don't want to just surface a keyword and say 'We found a keyword,'" LeGare says. "We want to surface a keyword and tell you what that means to your company in context." When looking at a customer tweet, for instance, Attensity's Semantic Annotation can examine the relationship between a sentence's different parts of speech and read it to recognize the overall sentiments.

By working with data scientists, sales and marketing teams can learn about trends in customers' attitudes and thought. Companies trying to determine why customers might be abandoning their products for those of a competitor can take quicker action using the technology.

Fern Halper, research director for advanced analytics at TDWI, notes that other companies probably offer comparable tools, but singles out Attensity as a noteworthy competitor. "Text analytics is a better method than search for gaining insights in text," she says, "and Attensity has some very interesting approaches that they are using to do this."

The product runs on an SDK engine, which can be dropped into an existing workflow, so customers who already have large datasets in place can use the technology without worrying about integrating it into their CRM systems.

LeGare says that the cost will be determined on a case-by-case basis, depending on the company's goals and the magnitude of the implementation.

Page 1
To contact the editors, please email editor@destinationCRM.com
Every month, CRM magazine covers the customer relationship management industry and beyond. To subscribe, please visit http://www.destinationCRM.com/subscribe/.
Related Articles
Medallia's text analytics solution brings unstructured data analysis to organizations.
The solution fully unifies structured survey data and unstructured text.
Social listening tool lets the technology product and service provider focus on actionable data.
New version of Attensity Analyze VOC solution brings more data to the discovery process.
Attensity partners with Informatica and Tableau Software to deliver sophisticated analysis of unstructured data within existing business applications and workflows.
Popular Articles

Home | Get CRM Magazine | CRM eWeekly | CRM Topic Centers | CRM Industry Solutions | CRM News | Viewpoints | Web Events | Events Calendar
DestinationCRM.com RSS Feeds RSS Feeds | About destinationCRM | Advertise | Getting Covered | Report Problems | Contact Us