eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
With predictive analytics at the core of many marketing initiatives, software developers are adding predictive capabilities to text mining so that enterprises can act on information in unstructured as well as structured data.
SPSS Inc. announced last week Predictive Text Analytics, a packaged solution that combines technology from the Chicago-based data mining software companys Clementine data mining tool and LexiQuest line of linguistics-based text extraction applications. SPSS purchased LexiQuest Inc. last February. SPSS joins a short list of vendors in this field that includes SAS Institute Inc., in Cary, N.C., which offers a text mining product called Text Miner as an add-on to its core Enterprise Miner data mining application.
A focus of both products is customer analytics, also known as analytical CRM (customer relationship management). SPSS Predictive Text Analytics combines text mining and data mining to analyze customers verbal responses to find reasons behind behavior patterns.
The SPSS solution includes LexiQuest Mine, a text mining tool; LexiQuest Categorize, a document categorization engine; and Text Mining for Clementine, an add-on to the Clementine workbench that applies classification, clustering and other data mining techniques to information gained through text mining. The software can then predict trends in customer attitudes and behavior, and business conditions, officials said.
The University of Pennsylvanias Abramson Cancer Center is using Clementine and LexiQuest in combination to better analyze years worth of textual data to gain new insights into cancer diagnosis and treatment. “We took 10 years of [medical] journal articles and mined them for terms and syntax,” said Michael Liebman, the centers director of computational biology and biomedical informatics, in Philadelphia. That analysis helps the center find patterns in the data showing relationships among diseases, symptoms, treatment and other factors, Liebman said, patterns that would otherwise have to be subjectively discerned by researchers poring over the data.
Hewlett-Packard Co. is using SAS Enterprise Miner and Text Miner to accomplish similar goals. “We look at free-form notes [from] our … reps and merge that with structured customer data,” said Randy Collica, senior analyst with HPs Enterprise Systems Group, Americas, in Littleton, Mass.
Discuss This in the eWEEK Forum