"A tremendous percentage of the data that business people have to work with -- the great majority of it - is in textual form," says David Folger, a vice president with industry analyst firm Meta Group. "You really need some way of dealing with that information."
And meeting that business need helped the SAS Text Miner beat out its competitors.
SAS, a major business intelligence company based in Cary, N.C., grabbed the category's top spot with nearly half the vote, with Cognos' BI Series 7 software finishing second. Other finalists were Corvu Corp.'s HyperVu, dfPower Studio 5.0 from DataFlux Corp. (an SAS company) and digiMine Inc.'s Customer Segment Manager.
What sets the Text Miner apart is its ability to handle textual information. Businesses receive a seemingly endless volume of text-based information every day. It rolls into an office in the form of emails, resumes, memos, sales notes, customer feedback and medical records. In most cases, there simply aren't enough people on staff to comb through each document and put the information in spreadsheets and structured databases, especially in this economy with layoffs and cutbacks greatly outweighing any news of new hires.
That information is often lost in the flood. Wasted.
But SAS designed the Text Miner to find themes and concepts woven into large documents. It automatically groups documents into topical clusters and classifies the material into predefined categories. It's even designed to integrate textual data with structured data to enrich the company's business intelligence capabilities.
"Text Miner lets you find patterns in a document, like a fancy version of a search engine in some ways," notes Meta Group's Folger. "It doesn't handle [structured information] but it integrates with it, and the rest of SAS's product line does handle it... Text Miner brings the two worlds together."
Randy Collica, a senior business/data mining analyst with Hewlett-Packard, says Text Miner was able to help him search for themes and concepts in a wide range of documents. HP beta tested Text Miner, and while one division of the hardware giant already has bought it, Collica says at least one other division is considering buying Text Miner.
"It really seemed to work out fairly nicely," says Collica, who has used the data mining tool on and off for about a year and a half, mainly in his B2B and consumer-side call centers. "It can pick up different topics and different themes. I was trying to look for different types of product themes -- things related to storage. It was able to pick up information -- gigabytes and disc space -- and know that they're all part of a central theme."
Collica says he also used SAS Text Miner to check on customer satisfaction and nip complaints and problems in the bud.
"They were looking for customers who had been dissatisfied or had issues with a call center agent, and we were able to parse out which customers had problems," says Collica. "If you had to do it by hand, that's a lot of work. Using Text Miner, we could go through them quickly and put them in categories of issues that might need to be brought up quickly."