Publications

The newly formed expert group for JSR-247, Java Data Mining 2.0, held its kickoff meeting at Oracle Corporation in Burlington, MA September 29-30, 2004. Some of the features being considered for JDM 2.0 include: time series, transformations, ensemble models, apply for association, mining unstructured data such as text, model comparison, feature extraction, multi-target models, and extending the JDM web services for these and other features. In JSR-247, the expert group plans to further simplify mining for developers, producing �macro�-like interfaces for common data mining tasks. Those interested in joining the expert group may submit a nomination form at http://www.jcp.org/en/jsr/detail?id=247.

Java Data Mining provides a framework for performing common data mining operations, while facilitating the introduction of new or non-standard mining algorithms. By separating algorithm details from function level specification, JDM strives to make data mining accessible to mining experts and novices. JDM currently focuses on mining functions including classification, regression, clustering, association, and attribute importance (a.k.a. feature selection).

The Java Data Mining 1.0 final specification was posted August 18, 2004 and is available at http://www.jcp.org/jsr/detail/73.jsp. The expert group expects commercial implementations of the standard to be available by year end and is encouraging adoption by university-sponsored data mining toolkits.

To facilitate public exchange of ideas on JDM, a project "datamining" has been created on java.net at https://datamining.dev.java.net/, providing a discussion forum, announcements, and document sharing among Java Data Mining users.

See http://www.jcp.org/introduction/overview/index.en.jsp for an introduction to the Java Community Process and becoming a member.