Aligning different ontologies from similar (or same) domains is an active field of current research. There are various solutions which process and analyze lexical, structural or semantic information to align ontologies. However, there are few solutions that focus on interpreting the concepts that entities are presented with and using them in relation to the semantics implied in an ontology. In this paper, the prototype (OACLAI) is presented to tackle this by combining lexical analysis with consequences from reasoners which reflect the semantics implied in an ontology. We evaluate OACLAI over the four real ontologies and compare it against the seven solutions. The experiments show that the accuracy of OACLAI is higher than those of others on average.