September 30, 2002–machine learning framework, now available from Wolfram Research, Inc., is a flexible Mathematica application package for business and engineering professionals who want to extract computational models from data.
With optimized, fuzzy logic-based machine learning methods and algorithms, machine learning framework can be a powerful tool for all types of data mining and machine learning applications:
- Media experts can extract features from images and signals to detect defects.
- Process engineers can model chemical, metallurgical, or other continuous processes to improve output.
- Process automation system developers can control complex manufacturing machines.
- Marketing professionals can create profiles to improve the positioning of their products and services.
- Financial engineers can analyze market characteristics to calibrate models for financial instruments.
All methods in machine learning framework are implemented in C++ and are seamlessly integrated into Mathematica‘s intuitive computation, visualization, and programming environment. Methods are highly parameterized to ensure maximum flexibility and performance in terms of computational speed and accuracy. Results can be easily visualized using the Mathematica front end and can also be modified using Mathematica programming to fine-tune the models.
Knowledge engineers and machine learning experts who want to customize, configure, and integrate their own machine learning arrangements will benefit from machine learning framework‘s open architecture and Mathematica‘s high-level programming environment.
machine learning framework‘s online user’s manual, which aids with function description and command reference, is fully integrated into the Mathematica Help Browser for convenience and ease of use. Also available are online notebooks with descriptions of all methods and templates for special predefined machine learning tasks.
machine learning framework is published and supported by Software Competence Center Hagenberg (SCCH) GmbH and uni software plus GmbH.
Additional information is available.