一种基于有向无环图的本体内聚度度量方法

An approach for ontology cohesion metrics based on directed acyclic graph

目前,随着本体的广泛使用和快速发展,本体在结构与语义上变得越来越复杂。如何对本体的质量进行评估成为本体构建和重用的主要问题。在本体构建过程中,对本体进行评估有利于对本体进行重构和优化,以构建高质量的本体。在本体重用过程中,可以帮助用户在候选本体集中选择最优结构的本体。提出一种基于有向无环图(DAG)的本体内聚度度量方法,首先依据有向无环图的结构提出一组本体内聚度度量指标;然后根据已有的度量验证框架对其进行验证,说明度量指标在理论上有效;最后使用经典本体数据集进行实验,说明所提出的本体内聚度度量方法的合理性和有效性,有利于本体的构建和重用。

With the widespread use and the rapid development of ontologies, the structures and se- mantics of ontologies are becoming more and more complex. The evaluation of the qualities of ontologies becomes a major problem for ontology construction and reuse. In the process of ontology construction, ontology evaluation is beneficial to the reconstruction and optimization of ontologies of high quality. In the reuse process, ontology evaluation can help users to select the optimal structure among the candidate ontologies. In this paper we first propose a set of evaluation metrics to measure the cohesion of the on- tologies based on directed acyclic graph （DAG）. Then we validate the theoretical effectiveness of the proposed metrics under the frameworks for software measurement validation. Finally, we use a set of classical ontologies to do experiments and the results demonstrate that the proposed metrics are reasona- ble and effective. It is conducive to the ontology construction and reuse.