基于地理本体的综合语义相似度算法

发布时间：2018-11-23 08:23

【摘要】：顾及到地理领域语义相似度计算模型考虑因素过于单一、主观性较强等问题,针对本体模型的结构特点,提出一种计算节点密度的新方法,并从模型概念间的关系类型、节点密度、节点深度等方面分析本体概念相似度的计算,将其归并为距离因素.基于本体层次网络结构计算语义信息量,该方法不依赖于专家经验,具有客观性.结合语义距离、信息量、属性等影响相似度的因素,提出一种计算概念问语义相似度的综合算法,该算法考虑到不同的影响因子在语义相似度计算中的重要程度不同,从而赋予地理本体关系不同的权值.通过对土地利用分类中实体的语义相似度进行实例验证,表明提出的算法能有效改善语义相似度计算的准确性和有效性,能够获得更符合认知的信息检索结果.
[Abstract]:In view of the fact that the semantic similarity calculation model of geographical domain is too single and subjective, a new method to calculate node density is proposed according to the structural characteristics of ontology model, and the relationship types between the concepts of the model are analyzed. The similarity calculation of ontology concept is analyzed in terms of node density and node depth, which is combined as distance factor. The semantic information is calculated on the basis of ontology hierarchical network structure. This method is not dependent on expert experience and is objective. Considering the factors affecting similarity, such as semantic distance, information content and attribute, a synthesis algorithm is proposed to calculate the semantic similarity of concept question. The algorithm takes into account the different importance of different factors in semantic similarity calculation. Thus the geographical ontology relationship is given different weights. An example is given to verify the semantic similarity of entities in land use classification. The results show that the proposed algorithm can effectively improve the accuracy and effectiveness of semantic similarity calculation and obtain more cognition-consistent information retrieval results.
【作者单位】：兰州大学资源环境学院;西北民族大学土木工程学院;中国科学院寒区旱区环境与工程研究所;
【基金】：国家自然科学基金项目(41271360,41571031) 甘肃省高等学校科研项目(2015B-003)
【分类号】：TP391.3

【相似文献】