Work

Optimizing Ontology Alignment via Multi-Objective Co-Evolutionary Algorithm

Year: 2026
Type: conference-paper
Abstract: Ontology Matching can bridge the gap between two heterogeneous ontologies, promoting the co-operation of the information systems built upon these ontologies. To improve the quality of matching results, this paper proposes a Multi-Objective Co-Evolutionary Algorithm (MOCEA) that effectively balances the convergence and diversity of pareto front by collaborating between two populations using different evolutionary strategies. Moreover, to accurately identify the timing of population interaction, this paper presents an improved population diversity analysis strategy. By eliminating a certain proportion of outliers, it achieves a more robust estimation of population diversity, thereby reducing the interference of extreme values. Experimental results show that the proposed algorithm can generate highquality matching results in the complex ontology matching task, significantly outperforming advanced matching methods.
Language: English
Open Access status: hybrid
OpenAlex ID: W7202344014
FWCI:
Cites: 7
Cited by:
Related to: 10