Formalization of Unstructured Content to Semantic Form.
Sandi Pohorec, Milan Zorman · Advances in Databases and Information Systems · 2011
This paper will present an approach for knowledge extraction from unstructured content. Unstructured content refers to information syndicated online, with little structure. The information extracted from these content sources is semantically tagged and stored in the form of semantic graphs. The approach is upgraded with the transformation of semantic graphs to a relational database. The goal of the content acquirement is to build a repository of formal knowledge, similar to DBpedia, for content in Slovene language. The formalized content will use the database as the permanent storage and would be provided in various formats suited for machine processing (RDF dump, API, SPARQL endpoint).