The XELOPES library expands the "emerging" OMG Common Warehouse Metamodel (CWM) Standard and at the same time represents one of its first implementations. Like the CWM it is specified completely in UML and thus platform independent. Implementations for Java, C++ and C# as well as CORBA and web service interfaces are presently available. A universal mining input stream concept allows the library to be applied to various data sources - from the main memory to files to databases. It is easy to program your own data access classes. That makes XELOPES completely independent of both programming language and data source.
By nature of its construction the XELOPES library is completely compatible with the CWM standard. The PMML data exchange format is extensively supported. Other supported BI standards are JMI and JOLAP. There are connectors for OLE DB for data mining as well as to different popular data mining libraries.
| Infoflyer XELOPES Library (en) Flyer (englisch) zur prudsys XELOPES Bibliothek geschätzte Downloadzeit ISDN (~ 64kB/s): 30 Sekunden | 178kB
| |
| The XELOPES Library: Embedded Data Mining Dr. Michael Thess, Presentation at Data Mining User Event Days 2003 (in German only) geschätzte Downloadzeit ISDN (~ 64kB/s): 17:4 Minuten | 6MB
| |
| XELOPES Basket Algorithm Benchmarks Benchmarks of Basket and Sequence Analysis Algorithms of XELOPES Java Version geschätzte Downloadzeit ISDN (~ 64kB/s): 16 Sekunden | 96kB
| |