基于形式概念分析与统计理论的本体构建模型

Ontology construction model based on formal concept analysis and statistical theory

为了解决目前本体建立技术严重依赖领域专家的参与、不够客观的现状,提出将形式概念分析与统计理论相结合的本体构建模型。通过利用统计算法对术语在文件中出现概率进行统计,根据设定的阈值对术语进行抽取,同时对术语出现的文件进行记录,进而将以上所得的知识组织成具有层次关系的概念格,最后将概念格映射为领域本体。实践证明,利用该模型建立的领域本体具有较好的客观性,同时能极大提高本体搜索的效率。

In order to solve that the current ontology technologies which mainly depended on the participation of experts in the field and were not objective enough, the thesis presented an ontology setting up technology with a combination of concept lattice and statistical study. First, counted the appearance probability of terms in the document by statistical arithmetic. Second, set up threshold of terminology extraction, noted the records of the files in which terminology appeared, and organized the concept lat- tices of knowledge organization-level relations. Finally, transformed the concept lattice into ontology. Practice shows that this method of setting up ontology has good objectivity, and can greatly improve the efficiency of ontology search.