Eng. Nazeer AlBokai
Syrian Virtual University
Bassel Alkhatib
Damascus University
Khaled Omar
Damascus University
ABSTRACT
Ontology is the philosophical study of being. It investigates what types of entities exist, and how they are related to one another on the most fundamental level, ontology is the skeleton of the semantic web, many algorithms developed to make the integration of existing ontologies on the world wide web , that these ontologies belong to different domains ,and its mandatory to benefit from these ontologies to integrate these ontologies based on domain, we propose a new automatic algorithm for ontologies integrations based on domain, and this algorithm will generate two types of semantic relations between ontologies Is_A relations and Part_Of relations, the proposed algorithm will use word embedding technology to generate Is_A relations, we have used PubMedBERT which is pre trained model from scratch using abstracts from full-text articles from PubMed Central, and wordNet to generate Part_Of relations, we develop this algorithm on the ontologies in medical domain based on the popularity of ontologies in this domain, we have tested our algorithm the accuracy rate was 85 %.
Keywords: Medical Ontologies, Semantic Web, Word Embedding, Ontology Integration, LLM.
