Probabilistic ontologies and probabilistic ontology learning: Significance and challenges

摘要

Human knowledge is limited therefore some information is incomplete or contradictory. When we develop an ontology, using an automatic ontology learning system or by human, with such information, the ontology would be inconsistent or we need to manage uncertain information. In non probabilistic approach, system discovers inconsistencies and then eliminates some parts of ontology to make it consistent. On the other hand, in probabilistic approach, discrepancies are adapted in the ontology.