Matching large-scale biomedical ontologies with central concept based partitioning algorithm and Adaptive Compact Evolutionary Algorithm

As a unified model for describing biomedical knowledge, a biomedical ontology is of help to solve the issues of data heterogeneity in different biomedical databases. However, these ontologies might model same biomedical knowledge differently, yielding the heterogeneity problem. To address the biomedical ontology heterogeneity problem, it is necessary to match the heterogeneous concept pairs between two ontologies. How to reduce the computational complexity is a challenging problem when matching large-scale biomedical ontologies, which directly affects the matching efficiency and the alignment’s quality. To face this challenge, this work proposes a large-scale biomedical ontology partitioning and matching framework. In our proposal, a central concepts based ontology partitioning algorithm is first used to divide the ontology into several disjoint segments, which borrows the idea from the social network and Firefly Algorithm (FA). The proposed algorithm is able to partition the ontologies with low computation complexity, and at the same time, ensure the semantic completeness and the decent scale of each segment. Then, an Adaptive Compact Evolutionary Algorithm (ACEA) based matching technique is utilized to determine the ontology segment alignments, which can efficiently match the similar ontology segments. The experiment utilizes the biomedical testing cases provided by Ontology Alignment Evaluation Initiative (OAEI) to test our approach’s effectiveness, and the experimental results show that the alignments obtained by our method significantly outperforms the state-of-the-art biomedical ontology matching techniques.