Abstract
With the widespread application of ontology in the fields of information retrieval, artificial intelligence etc, the concept similarity computation of different ontologies has been a research hot spot. At present, most research on concept similarity computation is based on ldquois ardquo relation between concepts, however, it does not utilize the concept semantic information completely, and the accuracy of calculating results is very poor. To solve this problem, a novel semantic similarity algorithm based on OWL ontologies is presented: Firstly, the method parses OWL ontologies, and then translates OWL ontologies into RDF triples, finally, utilizes an improved dynamic adjustment of the semantic weight of OWL constructors to calculate the semantic similarity. The experiment shows that the algorithm can obtain more accurate result.