With SPSS Predictive Enterprise Services, your organization can apply the same standards used by other business-critical IT practices to predictive analytics. As a result, you’ll experience lower total IT costs, enhanced satisfaction on the part of analysts and business managers, and increased confidence in the security of your organization’s valuable analytical assets.
SPSS Predictive Enterprise Services uses an advanced services-oriented architecture (SOA). By exchanging service-based messages between applications, SOA makes it easier to integrate disparate systems and applications, resulting in faster—and less costly—deployment of applications.
Analysts access the SPSS Predictive Enterprise Services repository through the user interface they’re familiar with—either the Clementine or the SPSS user interface. This increases their satisfaction with the process while reducing the installation burden for your IT staff.
The Predictive Enterprise Repository works with Microsoft® SQL Server™, Oracle®, or IBM® DB2® databases. Your IT staff can:
In addition, SPSS Predictive Enterprise Services efficiently supports other processes.
Configuration of multi-part processes. An analytic process often consists of several components. For example, scoring a model typically consists of a task to extract and prepare data, a task to actually score the model, and a task to write/transfer scores to the appropriate location for them to be used. Different parts of this process may involve the use of different applications and tools. Through the Predictive Enterprise Manager component, SPSS Predictive Enterprise Services provides a visual workflow enabling a process to be built from different components such as Clementine streams, SPSS syntax, SAS® code, command-line executed scripts, or other elements. Each step can be individually parameterized to tailor the process for execution using analytical or operational systems.
SPSS Predictive Enterprise Services provides a visual workflow, enabling a process to be built from a number of different components.
Click to enlarge
SPSS Predictive Enterprise Services can configure processes for:
Scheduling and notification. Many analytic tasks can be viewed as routine, in the sense that an organization would like to be able to perform them at set times (for example, daily, weekly, or monthly). A good example of this is re-scoring a churn model. Predictive Enterprise Manager allows jobs to be scheduled to run at various frequencies. Automatic notification allows key individuals or groups to be kept informed of the status of these automated processes.
Scheduling regular processes ensures that IT resources are used more efficiently, controls the peak-time load on your systems, reduces downtime for your analysts, and generates results more quickly.
Regularly scheduling an automated model building, validation, and scoring process can provide the benefits of “self learning” model building with the assurance that your new model will always be an improvement upon the old one.
If you use data mining or rules-based models to make key decisions, you may need to explain the basis for these decisions to satisfy compliance for Sarbanes-Oxley, Basel II, and other regulations. This executive brief explains how you can address regulatory compliance by using proven analytics management technology to automatically audit your predictive models, track your analytical assets over time, and centralize and secure predictive models. By doing so, you will be able to demonstrate compliance while protecting your analytical assets.
>Download this white paper
Predictive Analytics
can make your organization
more
successful
Key Products and Services
Resources
Buy
Demos and Downloads
