Abstract
Simple SummaryBiomedical ontology matching is a large-scale multi-modal multi-objective optimization problem with sparse Pareto optimal solutions. To effectively address this challenging problem, this paper proposes an adaptive multi-modal multi-Objective Evolutionary Algorithm. First, a novel multi-objective optimization model is constructed to simultaneously optimize both the alignment’s f-measure and its conservativity. Then, a problem-specific algorithm is presented, which uses the guiding matrix to adaptively guide the algorithm’s convergence and diversity in both objective and decision spaces. The experimental results show that our approach is able to effectively solve the biomedical ontology matching problem and to provide more options for decision makers.To integrate massive amounts of heterogeneous biomedical data in biomedical ontologies and to provide more options for clinical diagnosis, this work proposes an adaptive Multi-modal Multi-Objective Evolutionary Algorithm (aMMOEA) to match two heterogeneous biomedical ontologies by finding the semantically identical concepts. In particular, we first propose two evaluation metrics on the alignment’s quality, which calculate the alignment’s statistical and its logical features, i.e., its f-measure and its conservativity. On this basis, we build a novel multi-objective optimization model for the biomedical ontology matching problem. By analyzing the essence of this problem, we point out that it is a large-scale Multi-modal Multi-objective Optimization Problem (MMOP) with sparse Pareto optimal solutions. Then, we propose a problem-specific aMMOEA to solve this problem, which uses the Guiding Matrix (GM) to adaptively guide the algorithm’s convergence and diversity in both objective and decision spaces. The experiment uses Ontology Alignment Evaluation Initiative (OAEI)’s biomedical tracks to test aMMOEA’s performance, and comparisons with two state-of-the-art MOEA-based matching techniques and OAEI’s participants show that aMMOEA is able to effectively determine diverse solutions for decision makers.
Highlights
Biomedical ontology is able to address the biomedical data heterogeneity issue and to bridge the semantic gap among multi-source and multi-modal biomedical gaps
We present a problem-specific adaptive Multi-modal Multi-Objective EAs (MOEAs) algorithm, which adaptively maintains several populations to execute the search process and utilizes the Guiding Matrix (GM) to adaptively guide the algorithm’s convergence and diversity in both the objective space and decision space
The proposed adaptive Multi-modal MultiObjective Evolutionary Algorithm (aMMOEA) is employed on three biomedical tracks provided by the Ontology Alignment Evaluation Initiative (OAEI); the results reveal that aMMOEA is able to effectively determine the diverse solutions for DMs
Summary
The biomedical ontology matching problem is a MMOP with sparse Pareto optimal solutions To address this problem, we present a problem-specific adaptive Multi-modal MOEA algorithm (aMMOEA), which adaptively maintains several populations to execute the search process and utilizes the Guiding Matrix (GM) to adaptively guide the algorithm’s convergence and diversity in both the objective space and decision space.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
