What is KnowledgeMiner?
KnowledgeMiner^® is a data mining tool that enables anyone to use its unique form of modeling to quickly visualize new possibilities. It is an artificial intelligence tool designed to easily extract hidden knowledge from data. It was built on the cybernetic principles of self-organization: Learning a completely unknown relationship between output and input of any given system in an evolutionary way from a very simple organization to an optimally complex one.

Why choose KnowlegeMiner?
The main advantages of the inductive KnowledgeMiner approach are:

• Only minimal, uncertain a priori information about the system is required. That means, even if you have no experience in modeling, data analysis or designing a neural network you will be able to model, analyze and predict complex relationships of nearly any kind of system.
  
• A very fast and effective learning process for a personal computer. That means you can solve problems on your desktop in a reasonable time which you may have never thought possible.
  
• Modeling short and noisy data samples. That means, you can deal with a problem as is and don't have to construct artificial conditions for your modeling method to get it work.
  
• Output of an optimally complex model. Generally you can be sure to get a model at the end of the automated modeling process which can be expected not to be overfitted. Overfitted models are not able to predict inherent relationships between variables.
  
• Output of an analytical model as a transparent explanation component. That means, you can evaluate the analytical model to explain the obtained results immediately after modeling.

KnowledgeMiner 5.0 works on three advanced inductive learning modeling algorithms:

Self-organizing Networks of Active Neurons (SONAN; also known as GMDH Networks)

• This method creates parametric time series models, static or dynamic input-output models and predictable systems of equations. Up to 500 input variables could be considered for model creation, whereby at least 6 data samples are needed for each variable. The network structure is not predefined. A generated report for a linear model as part of a system of equations may look like this:
  
  
  
  (more reading)
  

Self-Organizing Fuzzy Rule Induction

• Working much like Self-organizing Networks of Active Neurons, this method generates fuzzy rules from fuzzy or boolean data. Using fuzzy variables like negative, positive or medium, the generated rules are composed of several AND, OR, NOT operators, and they show natural language-like descriptive power:
  
  

  

  (more reading)

Analog Complexing

Analog Complexing is a multidimensional pattern search method that can be used for clustering, classifying, and predicting most fuzzy objects. For prediction, for example, it self-selects several similar patterns relative to a given reference pattern and then uses their known continuations to form a prediction for the reference pattern. (more reading)

Model Base

KnowledgeMiner has a built-in model base to store and access all generated models of a document. Every model can be activated easily by the 'Models' menu to show graphs, report, analytical model description, and to use it for prediction on new data within the program.

The power and the advantages of KnowledgeMiner, compared with statistics as well as with traditional neural networks, make it easy to use and rapidly applicable to a wide range of real-world problems, and characterize it as the most effective modeling and prediction tool available.

Application areas

KnowledgeMiner's algorithms can be used for different data mining tasks:

KnowledgeMiner Utilities

1. TransformModel

This tool resolves KnowledgeMiner's linear and nonlinear SONAN models, converts them into spreadsheet models, and, finally, implements them in MS Excel automatically for immediate use in various runtime environments including non-Mac platforms.

2. Business Intelligence Workflow Case Study

This is a live database marketing application that is about reducing costs of a company's ad campaigns. It puts together several parts of the knowledge discovery process like accessing selected data from the database, data preprocessing (categorical into numerical values conversion and missing values handling), dimension reduction, data mining and validation using KnowledgeMiner's modeling services, and model combination by a network of active neurons into an autonomously running workflow process. If you are running a small business and want to take advantage of state-of-the-art business intelligence solutions, but don't like to afford the huge licence and maintainance costs of major data mining or CRM suites plus the expert to run them, the framework presented in this study might be what you are looking for. Try it out! It's free. Platinum users can build similar solutions by theirselves or you ask for our support services.

Additional Information

• New features in version 5

• Examples included in the KM package

• Why Is Data Mining Needed?: The Theoretical Background of KnowledgeMiner

• Related links: A brief selection of links on data sets and data mining.

• References: A (not complete) bibliography of self-organizing data mining books and papers.