IBM Text Mining Upgrade Is Social Media Savvy

New taxonomies make predictive analytics hip to Twitter slang and vertical-industry jargon.

Doug Henschen, Executive Editor, Enterprise Apps

May 11, 2010

2 Min Read

Does your text mining tool understand emoticons? The latest IBM SPSS Modeler 14 Premium Edition does. It also understands other forms of Twitter, blog and wiki speak and generational slang. Announced today, the upgrade also makes sense of jargon in vertical industries covered by 180 new taxonomies added with the release.

IBM SPSS is not new to text mining. Rosetta Stone, the provider of CD- and online-language-learning courses, uses IBM SPSS software to analyze answers to open-ended questions in surveys of current and potential customers. Rosetta Stone and other customers, including Money Mailer and the Navy Federal Credit union, correlate text analysis with structured data such as product information and transaction records to drive decisions on advertising, marketing and product development.

The ability to understand online slang and emoticons will help the many companies that are now interested in detecting and tracking customer sentiment about their products and services as expressed through Twitter, Facebook and other social media outlets.

"Emoticons are now part of what people write, and they are very important because they convey emotions that text, by itself, does not convey," said Erick Brethenoux, predictive analytics strategist at IBM. "There are also many slang terms, acronyms and text messaging usages of terms that are gaining adoption as members of Gen-Y are becoming consumers and entering the workforce."

The 180 new vertical industry taxonomies help IBM SPSS Modeler 14 make sense of banking terms like "APR" or "overdraft" in banking or "CORBA," "HIPAA" or "pre-existing condition" in insurance. Life Sciences, Telco and Consumer Electronics are among the other industries represented.

The telco industry was an early adopter of sentiment analysis of social media, but this style of text mining is headed mainstream, according to Brethenoux, with retailers and consumer electronics companies now being frequent practitioners. By examining comments within text streams from Twitter and Facebook, these organizations are trying to spot influencers within social networks and understand what they are saying. This has a big impact on customer churn and adoption of new products and services.

"In the telco industry, influencers have more extensive friend-and-family networks, they make more calls, and, when cross referenced with active Web site users, you see that they visit more often to seek out information on products," Brethenoux said.

Once influencer profiles are developed, similar customers can be identified and their behaviors modeled and better understood. Marketing programs can be developed and tested in hopes of swaying influencer behavior. Goals typically include reducing churn and maximizing cross-sell and up-sell in the short term, and improving products and services in the long term.

IBM SPSS Modeler 14 is now shipping worldwide. The Modeler Professional edition handles data mining while the Modeler Premium edition addresses both data mining and text mining. Prices were not disclosed.

About the Author(s)

Doug Henschen

Executive Editor, Enterprise Apps

Doug Henschen is Executive Editor of InformationWeek, where he covers the intersection of enterprise applications with information management, business intelligence, big data and analytics. He previously served as editor in chief of Intelligent Enterprise, editor in chief of Transform Magazine, and Executive Editor at DM News. He has covered IT and data-driven marketing for more than 15 years.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like

More Insights