摘要
结合现有模式匹配算法和GML模式的特点,给出了元素与元素语义贴近度的定义,在通用匹配规则的基础上引入具体的匹配规则.针对匹配规则在计算语义贴近度中的作用定义了权值调配函数,并给出了关于元素对的匹配函数的定义和计算公式.实验表明,该算法能够有效发现地理信息模式中元素之间的相似关系,提高了元素之间匹配结果的准确度,对地理信息模式的整合有较大的帮助.
Combined with existing scheme matching algorithm and the characteristics of geography markup language (GML), elements and semantic similarity between elements were defined. On the basis of general matching regulations, specific matching rules were introduced. After reviewing some existing solutions to schema matching; the formula for calculating the semantic similarity was presented. The formula was based on matching criteria, so the effect of individual criteria in the formula needs to be thought over. Thus the weight adjusting function was adopted to control the effect of individual criteria in the formula. The experimental results of this algorithm testified that it can effectively discovery the semantic similarity of the schema elements, and it can improve the precision of the matching results between elements. It will be a strong auxiliary tool for information integration.
出处
《华中科技大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2006年第7期46-48,共3页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(60273076).
关键词
模糊集
模式匹配
语义贴近度
权值调配函数
fuzzy set
schema matching
semantic similarity
weight adjusting function