DFKI-LT - An Information Extraction Approach to the Semantic Annotation of Folktales
An Information Extraction Approach to the Semantic Annotation of Folktales
2 First International AMICUS Workshop on Automated Motif Discovery in Cultural Heritage and Scientific Communication Texts, Vienna, Austria, University of Szeged, Hungary, 10/2010
We propose an Information Extraction (IE) approach to the automated semantic annotation of folktales. We introduce and motivate the type of templates that we consider for encoding the (possibly underspecified) information extracted from textual and linguistic units of the tales. Each template is (possibly partially) instantiated on the base of a combined use of linguistic annotation and semantic resources. We opt for an incremental strategy: already instantiated templates can be further specialized on the base of subsequently instantiated templates. Once the full text has been processed, a round of specialization and of merging of the instantiated templates can take place.
Files: BibTeX, AMICUS-IE.pdf