结合元信息和本体论的特点,针对目前知识融合中存在的效率问题,提出了基于元知识描述和本体库表示的知识融合系统.元知识的抽取简化了知识转换的过程,基于本体库的遗传融合算法在知识内涵的层次上构建了新的解知识空间,并对解知识空间的结构演化进行了分析.建立了系统评估和参数校正的自适应机制,详细地描述了该系统层次结构及各组成模块的工作原理和实现方法.实例分析的结果说明了系统实现的可行性和有效性.
The earlier process of extracting meta-knowledge predigests the diversion among multi-source knowledge. An ontology based genetic fusion algorithm is introduced to form a new knowledge space of solution, applying a self-adaptive mechanism of performance evaluation and refinement of parameters to promote the evolutionary process. The working principle and implementation of system structure and its composition modules are described in detail. The feasibility is discussed through a case study.