For prices, click here.

Common modeling technologies and their applicability to different data set dimensions.

• spreadsheet like handling of data including simple formulas and cell references
• several built-in mathematical functions for extending the data basis
• opens ASCII text files
• creates automatically
  • linear or nonlinear static regression models by Self-organizing Networks of Active Neurons (SONAN)
    • multi-input/single-output models as well as multi-input/multi-output models (system of equations) available analytically and graphically
  • linear or nonlinear dynamic regression models by SONAN
    • time series models, multi-input/single-output models as well as multi-input/multi-output models (predictable system of equations) available analytically and graphically
  • for up to
    • 500 input variables
• enables background modeling
• Receiver Operator Characteristic (ROC) for evaluation of the classification power of generated models
• stores all created models in a model base dynamically
• all models can be used for status-quo or what-if predictions, classification or diagnosis problems within KnowledgeMiner
• copy (PDF file) of the book by Mueller/Lemke "Self-Organising Data Mining"
• AppleScript support for program-to-program communication, task automation, and knowledge discovery workflow across the system or a network (not available on Windows systems). Read the book AppleScript for Absolute Starters by Bert Altenburg. It is free (PDF, 896k).
  
• creates nonparametric prediction models for fuzzy objects by Analog Complexing, an advanced pattern recognition technology for evolutionary processes. A synthesis of different prediction models (SONAN-based and Analog Complexing-based) is now possible as a powerful way to increase prediction accuracy.
• provides Fuzzy Rule Induction as a third self-organizing data mining method for modeling, classification and prediction tasks
• for the first time, integrated noise filtering characteristics for a second level, on-the-fly model validation; supports evaluation if a model reflects a causal relationship or if it just models noise
• two new data mining algorithms: Analog Complexing based clustering and classification (n classes)
• explicit definition of exogenous and endogenous variables for creation of systems of equations and their what-if type prediction
• TransformModel for implementing models in Microsoft Excel

• Self-organizing Networks of Active Neurons that perform
  • Active Neurons selecting their input variables themselves
  • advanced network synthesis and model validation techniques to end up in a robust, optimal complex model
• integrated two stage model validation
  • 1. level: leave-one-out cross-validation driven model synthesis to avoid overfitted models (noise filtering)
  • 2. level: noise filtering characteristic of the first stage is applied to the final model to check if it really reflects some causal relationship (model evaluation)
• creation of a best and autonomous system of equations (a network of Self-organizing Networks of Active Neurons) that is ready for long-term status-quo and what-if predictions of the complete system; every system is available analytically (equations or rules) and graphically (system graph of the interdependence structure) for results interpretation
• Analog Complexing as a powerful pattern search technology to create cluster, classifications, or predictions for fuzzy processes (the most market processes e.g.) which other methods may be not appropriate for.
• Fuzzy Rule Induction from data to describe objects in a more natural language qualitatively
• explanatory power of any created model by default
• a model base to store all models and to keep connected information together
• completely autonomous modeling process that can work as background process on your computer saving your resources either by working simultaneously with the modeling process or, for larger problems, by running the process overnight
• allows for knowledge discovery workflow processing via AppleScript (Mac only)