SPSS Enhances Clementine for CRM, Fraud Detection
Clementine has been used as a complement to CRM for half a decade now—but SPSS recently taught it a range of new tricks.
SPSS Inc. this week announced the newest release of its Clementine data mining workbench, Clementine 10.
SPSS has marketed Clementine as a complement to CRM and other customer-centric applications for nearly half a decade now. In late 2001, for example, the company announced Clementine 6.5, which shipped with a first-ever Clementine Application Template (CAT) for CRM. And earlier that same year, SPSS launched its dedicated CustomerCentric Solutions division.
So the Chicago-based data mining stalwart has long been a fixture in the customer data marketscape.
This doesn’t mean that even so experienced a dog as Clementine can’t be taught a few new tricks, SPSS officials note. “Clementine 10 is a ... step forward in markedly enhancing the utility and power of data mining,” says SPSS spokesperson Marc Brailov. “It will elevate productivity and drive cost-efficiency throughout the data mining process. And it offers important features that will boost CRM, marketing, revenue assurance and fraud detection applications.”
For example, Brailov and other SPSS officials say, the revamped Clementine 10 has been tweaked not only to provide additional insight into CRM and other kinds of customer-centric data, but also to enhance the capabilities of fraud detection and revenue assurance applications.
Consider CRM, Clementine’s bread-and-butter app. In this case, SPSS officials say, the revamped Clementine incorporates new “feature selection” capabilities that help enhance the productivity of CRM and marketing analysts, especially for scenarios such as customer acquisition, cross- and/or up-selling, and customer retention. Feature selection produces more intuitive models that help improve customer insight and simplify operational deployment. Similarly, Clementine 10 ships with a new “anomaly detection” feature that helps shore up its fraud detection and revenue assurance capabilities. Anomaly detection can simplify analysis and scoring, improve insight, and—once again—facilitate the use of these insights in operational deployments, SPSS officials say.
Elsewhere, analysts can now rank and filter attributes in a number of different ways, which helps simplify the model building process, officials say. Clementine 10 ships with improved Excel support, too: analysts can now export data directly to Excel from the Clementine interface; what’s more, when analysts are importing data from Excel, they can specify worksheets and data ranges.
Finally, the revamped Clementine ships with a range of performance improvements, including in-database caching, database write-back with indexing, and optimized merging for joining tables outside of the database.
Stephen Swoyer is a Nashville, TN-based freelance journalist who writes about technology.