IBM's SPSS Update Blends Social, Business Data

Big Blue's new data mining and text analytics software culls information from Facebook and blogs to give companies a faster and more detailed understanding of their customers' needs.

IBM this week unveiled the latest update to its SPSS predictive analytics software, giving enterprise customers the ability to monitor and analyze social media data that's constantly being updated and shared via Facebook, Twitter, RSS feeds, blogs and wikis.

IBM SPSS Modeler promises to monitor changes in customer and employee sentiments, uncover deeper insights and predict the key factors that businesses need to grow their sales, retain top performers and quickly react to subtle changes in their industries or product categories.

The software's text analytics component identifies emoticons that are often included in a blog entry or tweet and can discern between different meanings or references for the same word. It also understands slang terminology and abbreviations so users get an accurate understanding of what people are saying about their companies and products in the vernacular that tweeters and bloggers use.

IBM (NYSE: IBM), which acquired SPSS for $1.2 billion last year, said customers such as Navy Federal Credit Union, Rosetta Stone and Money Mailer are already using SPSS Modeler Professional to make "faster and more personalized decisions" from information culled from a variety of data sources.

Companies large and small are spending billions on software tools that deliver real-time social networking analytics not only to track consumer behavior and sentiment but to offer discounts and promotions on the fly.

One of the biggest challenges for any company tracking their brand online is creating the taxonomies and definitions that fully encompass everything they sell and do, so whatever tracking or analytics software they're using knows what to look for out in the expansive social media world.

IBM thinks it's addressed most of those concerns by creating new semantic networks with 180 different vertical taxonomies -- covering everything from consumers electronics and life sciences to banking and insurance -- and some 400,000-plus terms, including more than 100,000 synonyms and brands.

Armed with the intelligence cobbled from Twitter, Facebook, industry blogs, news blogs and other online sites, companies can then merge this analytical data with structured data derived from customer e-mails or call center notes to build a complete view of, for example, a new product release to see what customers, partners and competitors are saying about the product.

Eventually, IBM sees customers using this data to more accurately predict demand for new products, quickly reverse poor policy or product changes and more effectively market and advertise to online shoppers.

According to the most recent survey by Forrester Research, interactive marketing spending will explode to more than $55 billion by 2014 and represent more than 20 percent of total advertising spend.

Also, the software gives companies a great tool to help monitor what the world is saying about its top competitors.

"Predictive analytics allows us to leverage unsolicited and unbiased customer feedback and strategically improve our business," Nino Ninov, vice president of strategic research at Rosetta Stone, said in a statement.

"We now can also monitor competitor and industry websites, including blogs and news feeds, and other publicly available textual information to maintain a current view and better understand how the public perceives our competition."

Larry Barrett is a senior editor at, the news service of, the network for technology professionals.

Tags: IBM, business intelligence, data mining, business intelligence software, analytics

0 Comments (click to add your comment)
Comment and Contribute


(Maximum characters: 1200). You have characters left.