| 基于图学习的本体概念相似度计算 |
| |
| 引用本文： | 高炜,梁立,张云港.基于图学习的本体概念相似度计算[J].西南师范大学学报(自然科学版),2011,36(4). |
| |
| 作者姓名： | 高炜 梁立 张云港 |
| |
| 作者单位： | 云南师范大学信息学院,昆明,650092 |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 根据边的类型、顶点深度、边的密度和强度以及边关联的两顶点的属性计算有向边的权重,通过图学习正则化模型得到优化函数.将本体结构图中每个顶点映射成一个实数,通过比较实数间的差值判断两概念的相似程度.实验表明该方法对于计算本体概念间的相对相似度是有效的.
|
| 关 键 词： | 本体 概念相似度 语义亏损 正则化模型 图的容量 |
Ontology Concept Similarity Computation Based on Graph Learning |
| |
| GAO Wei,LIANG Li,ZHANG Yun-gang.Ontology Concept Similarity Computation Based on Graph Learning[J].Journal of Southwest China Normal University(Natural Science),2011,36(4). |
| |
| Authors: | GAO Wei LIANG Li ZHANG Yun-gang |
| |
| Institution: | GAO Wei,LIANG Li,ZHANG Yun-gangSchool of Information,Yunnan Normal University,Kunming 650092,China |
| |
| Abstract: | To study semantic similarity of the concept is an important aspect in a knowledge representation and information retrieval.Compute the edge weights according to the type of edge,vertex depth,density and intensity of edges and attributes of vertices associated with edges.Obtain the optimization function using graph regularization framework.The function maps the vertices of graph to real numbers.Calculate the relative similarities of concepts by comparing the difference of corresponding values.Experiments sho... |
| |
| Keywords: | ontology similarity computation semantic loss regularization framework volume of graph |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《西南师范大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《西南师范大学学报(自然科学版)》下载免费的PDF全文 |
|
