C4.5: programs for machine learning
C4.5: programs for machine learning
An Algorithm for Multi-relational Discovery of Subgroups
PKDD '97 Proceedings of the First European Symposium on Principles of Data Mining and Knowledge Discovery
Data Mining with C4.5 and Interactive Cartographic Visualization
UIDIS '99 Proceedings of the 1999 User Interfaces to Data Intensive Systems
Hi-index | 0.00 |
Kepler is an extensible data mining platform that supports the entire knowledge discovery process from data access and preparation to analysis and visualization. One of its particular strengths is its open plug-in architecture, which allows third-party developers to easily integrate analysis tools and to import formats or preprocessing operators without the need to re-implement existing software. A large number of popular analysis algorithms can be used as Kepler plug-ins, including such classics as regression, decision trees, association rules, and clustering, as well as instance-based methods, Bayesian approaches, and subgroup discovery. Furthermore, Kepler is able to work with data that is stored in more than one table. Foreign links can be defined and used by several analysis techniques. For most of the tasks mentioned above, both single-relational and multirelational plug-ins are available. Kepler is scriptable, and thus, a good workbench for analysts and developers. Kepler employs a Java client and features a three-tier architecture that links to relational databases. The architecture allows specialized vertical data mining solutions to be constructed in domains such as marketing/finance, electronic commerce, and science/engineering. The analysis of geographically referenced data is now also possible through a link to the Descartes interactive geographical data exploration environment.