Biomedical ontologies have gained particular relevance in the life science domain due to its prominent role in representing knowledge in this domain. However, the existing biomedical ontologies could define the same biomedical concept in different ways, which yields the biomedical ontology heterogeneous problem. To implement the inter-operability among the biomedical ontologies, it is critical to establish the semantic links between heterogenous biomedical concepts, so-called biomedical ontology matching. Since modeling the ontology matching problem is a complex and time-consuming task, swarm intelligent algorithm (SIA) becomes a state-of-the-art methodology for solving this problem. However, when addressing the biomedical ontology matching problem, the existing SIA-based matchers tend to be inefficient due to biomedical ontology’s large-scale concepts and complex semantic relationships. In this work, we propose a compact firefly algorithm (CFA), where the explicit representation of the population is replaced by a probability distribution and two compact movement operators are presented to save the memory consumption and runtime of the population-based SIAs. We exploit the anatomy track, disease and phenotype track and biodiversity and ecology track from the ontology alignment evaluation initiative (OAEI) to test CFA-based matcher’s performance. The experimental results show that CFA can improve the FA-based matcher’s memory consumption and runtime by, respectively, 68.92% and 38.97% on average, and its results significantly outperform other SIA-based matchers and OAEI’s participants.

中文翻译：

---

一种用于匹配生物医学本体的紧凑萤火虫算法

由于生物医学本体在表示生命科学领域中的知识方面发挥着重要作用，因此它在生命科学领域中具有特别重要的意义。然而，现有的生物医学本体可以以不同的方式定义相同的生物医学概念，这产生了生物医学本体的异构问题。为了实现生物医学本体之间的互操作性，至关重要的是在异构生物医学概念之间建立语义联系，即所谓的生物医学本体匹配。由于对本体匹配问题进行建模是一项复杂且耗时的任务，因此群体智能算法（SIA）成为解决此问题的最新方法。但是，在解决生物医学本体匹配问题时，由于生物医学本体的大规模概念和复杂的语义关系，现有的基于SIA的匹配器往往效率低下。在这项工作中，我们提出了一种紧凑的萤火虫算法（CFA），其中人口的显式表示由概率分布代替，并且提出了两个紧凑的运动算子以节省基于人口的SIA的内存消耗和运行时间。我们利用本体比对评估计划（OAEI）的解剖轨迹，疾病和表型轨​​迹以及生物多样性和生态轨迹来测试基于CFA的匹配器的性能。实验结果表明，CFA可以分别将基于FA的匹配器的内存消耗和运行时间平均提高68.92％和38.97％，其结果明显优于其他基于SIA的匹配器和OAEI参与者。