Specifying parallel composition of matchers for ontology matching by using genetic algorithm

Ontology matching is the process of finding semantic relationships or correspondences between entities of different ontologies. It is a key challenge in the integration of heterogeneous applications. In this paper our focus is on aggregation methods that aggregate basic matchers in their parallel composition. Various aggregation methods are tested on two pairs of ontologies. Given the results, weighted aggregation method is the best method for aggregation in parallel composition of basic matchers. For each pair of ontologies, different assignment of weights of the basic matchers provides the best matching results so the equally weighted aggregation is not appropriate. This paper proposes using of genetic algorithm for determining the weights of basic matchers. The use of genetic algorithm accelerates the search of the ideal weights of each basic matcher for every pair of ontologies.