Application domain | Text categorisation by content and genre |
Source | LIMAS corpus of contemporary German |
Dataset size | 500 documents with 2000 words each, up to 75,000 facts per run |
-Data format | RIBL (Sets of ground facts in Prolog notation) |
Systems used | RIBL, LVQ (OLVQl), IBL, IEL-IG |
References | [1] |
Pointers | mathias.kirsten@gmd.de |