Explorative data mining with visual cluster analysis,
statistical profiling and segmentation

The Viscovery Profiler is a high-end solution for explorative data mining based on self-organizing maps (SOM) that provides tools for visual cluster analysis, statistical profiling and segmentation. By consolidating class information, cluster models can be designed as efficient classifiers, e.g., for business applications such as customer segmentation or technical tasks such as optical character recognition.

Viscovery Profiler key benefits

Compact data representation and powerful visualisation

• Compact representation and visualisation of large amounts of data from data marts and other external sources, based on leading commercial implementation of SOMs

• Modelling tool for generating high-performance segmentation models

• Accessibility of statistical information and original data (from any map partition) at any time

Interactive analysis, accurate classification

• Tools for defining clusters, segmentations and group profiles, and assigning operational measures to target groups

• Context-sensitive tools for statistical analyses (correlation matrix, PCA, histograms, box plots, scatter plots, etc.)

• Classification of new data records into segments manually or automatically

Predictive modelling without programming

• Full workflow support and administration of relevant parameters

• Integration of expert knowledge as segment-specific formulas

• Drag-and-drop reuse of completed Viscovery Profiler workflows at any time or for automatic execution

• Inline reporting functions

As a software module of the Viscovery Data Mining Suite, the Viscovery Profiler also provides the suite's general functionalities and benefits.

Viscovery Profiler features and functions

In contrast to traditional analytical systems, Viscovery Profiler provides an intuitive approach to data, so that users who do not have advanced statistical knowledge can easily understand and use the data. Target groups are easily chosen and assigned to campaigns and other measures, and the success of the actions can be evaluated in a separate workflow.

All activities, from importing the data to defining segments and measures, are guided by workflows. Any resulting segmentation model can be applied either interactively or in batch mode. In addition, the Viscovery One(2)One Engine supports real-time integration of Viscovery Profiler models with up to 100,000 decisions/second.

Statistical information, as well as the original data, are available at any time for each cluster or target group. Data records can be analysed in real-time and returned to the operational system together with the appropriate measures. Completed models can be modified and applied to new projects.

Viscovery Profiler is also available as a stand-alone product.