Release Notes: The latest version of the Weka machine learning library was added, together with some other small additions and extensions. Several minor bugs were also fixed.
Release Notes: This release provides 13 new operators, more than 10 bugfixes, and a lot of exciting new features like the new operator “Script” allowing for arbitrary Groovy scripts, improvements for the join and the aggregation operators, faster name-based access of attributes, and several usability enhancements.
Release Notes: This release provides 35 new operators, more than 30 bugfixes, and many new features. New Preprocessing Options: removal of duplicates, nominal value splits, data set union and superset, and others. New Learning Schemes: linear and quadratic discriminant analysis, fast large margin, and others. New Process Possibilities: FileIterator, MacroConstruction, SetData, ExceptionHandling, and others. Improvements include better handling of time zones, helper operators for attribute renaming, and others.
Release Notes: This release focuses on the most requested data analysis, ETL, and BI requirements. It provides more than 50 new operators and a lot of new features including upgraded data pivotings, new aggregation functions, and date and time handling. The usage of function-based attribute construction was simplified, and optimized wizards or new visualizations, including zooming and panning, were added. More powerful processes are now possible due to enhanced macros and the new result storage mechanism. This release also provides fixes for more than 30 bugs.
Release Notes: This is a minor bugfix and feature release. In particular, two errors that caused problems in the new parameter optimization wizard for string and integer parameters were fixed. New features include a new 64-bit version for Windows x64 systems, multiple groupings for aggregations, better support for date and time columns, the grouping and ungrouping of models, and an improved attribute subset preprocessing. Additionally, the runtime of several analysis schemes was drastically reduced, in some cases up to a factor of 13.
Release Notes: This is a major bugfix and feature release. The new features include a 64-bit version for Windows x64 systems, an improved and more comprehensive documentation of all operators inside the program, two different icon sets and look and feels, and a new and improved plotting facility including different shapes for plotter points. The most important preprocessing operators can now produce both views and materialized data changes, and several new preprocessing models were added. New wizards for parameter optimization were added. About 80 bugs were fixed.
Release Notes: This release is mainly a bugfix release. In particular, two errors which caused unusually high memory consumption for some data mining processes were fixed. More precise data representations are also used again, since too many rounding errors occurred. In total, about additional 30 bugs were fixed compared to version 4.1beta and it is therefore highly recommended to change to this version instead. Additionally, some improvements for plotting and some new operators were added.
Release Notes: This version provides about 30 bugfixes, together with a set of new features, including a huge reduction of needed memory, new visualization techniques for association rules, and improved rule and tree learners. The connection to databases was improved, and about 10 new operators were added. The bugfixes include the wizard and building block bug introduced in version 4.0 and several database connection bugs (especially for connections to the Microsoft SQL Server 2005 database).
Release Notes: This release supports workspaces for different projects. The training and the test data no longer need to have exactly the same structure. The operator PerformanceEvaluator is now divided into smaller task dependent operators. Rule learners are able to learn from numerical data. Several new visualization techniques and plotters were added. This release also fixes many bugs, including a name bug for attribute names with different cases and a bug in the Anova calculation operators.
Release Notes: Among some new features and operators, this release offers many bugfixes related to GUI issues and the external operators (Weka and LibSVM). Probably the most important enhancements were done for the decision tree and rule learning algorithms and their visualizations. New operators for association rule mining also exist. This release also contains the latest Weka version and some new operators.