BOEMIE System
Jump to navigation
Jump to search
See: BOEMIE Project, Research System, Ontology Evolution.
References
2011
- (Karkaletsis et al., 2011) ⇒ Vangelis Karkaletsis, Pavlina Fragkou, Georgios Petasis and Elias Iosif. (2011). “Ontology Based Information Extraction from Text.” In: Knowledge-Driven Multimedia Information Extraction and Ontology Evolution. Springer. doi:10.1007/978-3-642-20795-2_4
- QUOTE: In BOEMIE system (Castano et al., 2008), the core idea is the bootstrapping of ontology evolution in the framework of ontology-based information extraction. The extraction process is layered, having in the first layer the identification of ontological concepts and their relations that can be attributed to text segments. In the next layer more composite concepts of higher level and relations among them are generated, based on the previously extracted concepts, using an inference mechanism. In contrast to the lower-level concepts of the first level, the higher-level concepts usually cannot be mapped to textual fragments. For example, assume that the instances referring to an athlete, a sport and a tournament were extracted in the first layer. The inference mechanism relates the instantiations of these concepts, generating a higher-level concept according to the domain ontology. The ontology evolution task of BOEMIE system can be roughly distinguished into two branches: ontology population and ontology enrichment. The procedure of ontology population adds new individual entities to the ontology by accounting disambiguation and consistency maintenance issues. The domain knowledge is extended by the addition of new concepts and relations that are obtained through the process of ontology enrichment.
2008
- (Castano et al., 2008) ⇒ Silvana Castano, Sofia Espinosa, Alfio Ferrara, Vangelis Karkaletsis, Atila Kaya, Ralf Möller, Stefano Montanelli, Georgios Petasis, and Michael Wessel. (2008). “Multimedia Interpretation for Dynamic Ontology Evolution.” In: Journal of Logic and Computation, 19(5).