摘要
句子语义距离计算在许多智能系统中发挥着重要的作用.针对句子语义知识的表达方式、语义框架,提出一种基于语义同构的语义距离算法,通过计算代表句子语义的两个语义框架的最大同构子框架和最大包容子框架,计算句子的语义距离,并且通过动态调整句子成分的权重,提高算法的有效性.实验证明,该算法可以使语义距离计算精确率达到97.7%,与现有方法相比具有更高的精确度.
The calculation of the semantic distances between sentences plays an important role in many intelligent systems. According to the semantic representing system-semantic framework, a semantic - isomorphism-based semantic distance algorithm is proposed. By calculating the maximum common sub - framework isomorphism and maximum containing sub-framework of the two semantic frameworks, the algorithm calculates the sentence semantic distance, and by dynamically adjusting the weights of the sentence ingredients, the validity of the algorithm is improved. Experiments show that the precision of the algorithm achieves 97.7 %, higher than existing algorithms.
出处
《山东师范大学学报（自然科学版）》
CAS
2008年第3期50-53,共4页
Journal of Shandong Normal University(Natural Science)
关键词
语义距离
语义框架
语义同构
动态权重
semantic distance
semantic framework
semantic isomorphism
dynamic weight
