Manual creation of ontologies is a time-consuming, costly and complicated process. Consequently, over the past two decades a significant number of methods have been proposed for (semi)automatic generation of ontologies from existing data, especially textual ones. However, ontologies generated by these methods usually does not meet the needs of many reasoning-based applications in different domains. This study is an effort towards reusing the freely available knowledge in Linked Open Data as background knowledge beside text in order to improve the results of ontology learning from text in terms of multilingual making of ontology terms, classification of dangling instances, recommending appropriate intensions for ontology concepts, concept hierarchy enrichment and extracting non-taxonomic relations. The preliminary experimental results show the importance of the proposed approach through the achieved improvements in most of the objectives mentioned. Additionally, Deep Learning approaches are now taking off in various fields, including Ontology Learning from text. Some recent efforts report improvements in ontology learning from text by using Deep Learning approaches. Accordingly, the potentials of Deep Learning approaches in regard to ontology learning will be addressed shortly in this talk.
18.04.2019 - 10:15