Enhancing ontology alignment through a memetic aggregation of similarity measures

Interactive ontology matching based on partial reference alignment

“…In this study, an ontology O is defined as 5-tuple (C, P, I, Λ, Γ) [7], where C, P, I, Λ, Γ are referred to the set of classes, properties, instances, axioms and annotations respectively. In addition, an ontology alignment A between two ontologies is a correspondence set.…”

“…Therefore, it is able to solve the problem of ontology matching in a more efficient way. Actually, HEA-based ontology matching techniques perform better in determining ontology alignments than state-of-the-art ontology matching systems do [7,8,9]. However, since it is necessary to execute the ontology matching within a specific run time, aparting from the quality of alignments, the execution time and main memory consumption are essential too.…”

A self-adaptive Multimeme Memetic Algorithm co-evolving utility scores to control genetic operators and their parameter settings

“…A result of note from this work was that the choice of local search mechanism was found to be more critical to the overall performance of an MA than the choice of underlying population-based search strategy. Acampora et al [1] tested a large number of MA configurations applied to ontology alignment, based on the issues outlined in Nguyen et al [45], with the best MA found to be competitive with state-of-the-art ontology alignment systems.…”

OMI-DL: An Ontology Matching Framework

“…Some of the techniques include similarity flooding [18], [19], [20], coefficient computation [11], [21], graph matching [22], [23], formal concept analysis [24], machine learning [25], [26], Bayesian decision theory [7], hybrid methods [8], [27], [28], Markov networks [9], optimization techniques [29], [30], or reasoners [10], [11], [12], [10], [31].…”