基于改进粒子群优化算法的Ontology划分方法

为解决规模巨大的Ontology难以使用的问题,提出了一种基于改进粒子群优化算法的Ontology自动划分方法.根据Ontology划分的要求,将概念落入某个子Ontology的概率作为粒子的速度,而将概念落入的子Ontology编号组成的数字串作为粒子,设计了粒子群优化算法的适应度函数,并给出了Ontology划分算法的具体步骤.最后进行了相关对比实验,结果表明,该方法具有比其它方法更好的划分效果.

Ontology Partition Method Based on Improved Particle Swarm Optimization Algorithm

In order to overcome the difficulty in the use of huge Ontology,a method of automatic Ontology partition is proposed based on the improved particle swarm optimization algorithm.In this method,the probability of the concept to fall into a certain sub-Ontology is taken as the particle speed according to the Ontology partition request,and the digital string of the number of the sub-Ontology is taken as the particle.Afterwards,the fitness degree function of the particle swarm optimization algorithm is designed,and the concrete steps of the Ontology partition are presented.According to the correlation contrast experiment,it is finally found that the proposed partition method is of better division effect than the other methods.