New release: on Tuesday, July 31st, 2007, Rapid-I released a new version of the open-source data mining software RapidMiner 4.0 (formerly: YALE).
Because of legal issues, Rapid-I decided to change the name of YALE. The name RapidMiner was chosen for its fit to the company name Rapid-I and to the naming scheme of the line of products planned by Rapid-I for the future. Only the name changes. Everything else stays the same. RapidMiner / YALE will remain open-source software under GNU GPL and available to end-users free of charge and is also available under a commercial OEM license.
Changes from RapidMiner 4.0beta2 to RapidMiner 4.0
General Improvements:
RapidMiner now supports workspaces for different projects. The default workspace must be set during the first start of RapidMiner.
Training and test data do no longer need to have exactly the same structure. Changing the order of nominal values often caused problems for past versions.
The operator PerformanceEvaluator is now divided into smaller task dependent operators allowing for additional checks.
New compatibility checks for prediction models between training and application data.
A filter for the New Operator tab is added.
Added learning for numerical attributes for rule learners.
Improved the visualization for almost all performance criteria.
Added 3D visualization of confusion matrix
Added automatical (averaged) ROC curve visualization for AUC criterion.
New plotter: Deviation.
amongst others
Screenshots
New Operators:
Performance
ClassificationPerformance
BinominalClassificationPerformance
RegressionPerformance
UserBasedPerformance
SingleRuleWeighting
MPCKMeans
Bugfixes:
This release contains many bugfixes including the removal of unnecessary parameters from RandomForest, name mixing bug for attribute names with different cases, a bug in the Anova and t-test calculation operators, a bug preventing the visualization of cluster model graphs, a bug for discretization operators inside of looping chains, a bug in rule learners preventing greater equal conditions, a bug during the saving of neural network models amongst others.