Company

New Platform Support and Features in machine learning framework 1.5

April 27, 2006--In response to increasing user demand, the latest version of machine learning framework 1.5 is now available not only for the Windows platform but for Linux and Mac OS X as well. This Mathematica application package also adds new features for real-world, multi-method data-mining projects--including neural networks and quadratic regression models--and improves existing methods and cross-validation techniques.

machine learning framework is a complete solution for business and financial engineers, process and manufacturing engineers, quality assurance professionals, and all experts who want to extract computational models from data. With the unprecedented power of Mathematica's programming environment, machine learning framework is relied on by industry leaders to customize and configure machine learning solutions, search for quick modeling capabilities, and to predict and control input-output relations. Mathematica's extensive connection technologies make it easy to integrate machine learning framework into runtime environments such as process automation and manufacturing management systems.

In addition to new platform support, machine learning framework 1.5 includes the following new features and enhancements:

  • New data manipulation, regression, and classification algorithms, such as for neural networks and quadratic regression models
  • Improved ridge regression and linear regression trees (LIRT)
  • Better visualization of linear regression models, as well as LIRT and ID3 trees
  • Increased performance of cross-validation techniques
  • Statistics now available for attributes of all major algorithms
  • Supplemented error statistics
machine learning framework 1.5 can be downloaded from the Wolfram web store. Developed and supported by Software Competence Center Hagenberg (SCCH) GmbH, it requires Mathematica 5.2 or higher and is now available for Windows, Linux, and Mac OS X. Upgrades and trial versions are also available.

See the product pages for more information about machine learning framework.