Inductive Lexical Learning of Class Expressions

摘要: Despite an increase in the number of knowledge bases published according to Semantic Web W3C standards, many those consist primarily instance data and lack sophisticated schemata, although availability such schemata would allow more powerful querying, consistency checking debugging as well improved inference. One reasons why are still rare is effort required create them. Consequently, numerous ontology learning approaches have been developed simplify creation schemata. Those usually either learn structures from text or existing RDF data. In this submission, we present first approach combining both sources evidence, particular combine logical with statistical relevance measures applied on textual resources. We perform experiment involving a manual evaluation 100 classes DBpedia 3.9 dataset show that inclusion leads significant improvement accuracy over baseline algorithm.