<$BlogRSDUrl$> Marcus P. Zillman, M.S., A.M.H.A. Author/Speaker/Consultant
Marcus P. Zillman, M.S., A.M.H.A. Author/Speaker/Consultant
Internet Happenings, Events and Sources


Saturday, July 03, 2004  

Text Mining Begins Digging Through Unstructured Data
http://snipurl.com/7gpp
A good 85% of an organization's knowledge is in the form of unstructured data. Easy to quantify, hard to find. "We are drowning in information but are starving for knowledge," says an R&D technical leader at Dow Chemical. But a new generation of text mining tools is allowing companies to extract key elements from large unstructured data sets, discover relationships and summarize the information. Dow, for example, uses ClearResearch to extract data from chemical patent abstracts, published research papers and the company's own files. And the University of Louisville uses SAS's Text Miner on text files, such as patient charts, and analyzes flat-file snapshots of billing and pharmaceutical databases as text, rather than as database entries. Researchers there have pinpointed certain medications that can prolong hospital stays for patients. Because of some limitations in the new software (such as understanding linguistics), text miners are still niche products, generally restricted to specific parts of an organization and requiring specialized analytical skills to implement and deliver truly useful information. It'll be awhile before they're commonly available. But some vendors are already incorporating text mining tools as a background function to improve the effectiveness of more familiar search or document management applications.

posted by Marcus Zillman | 4:10 AM
archives
subject tracers™