欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出了一种针对网页结构树的相似度计算方法,首先把网页标签结构表示成树,然后通过动态规划算法,使两棵树在每一层上最相似的儿子节点继续进行比较,而那些没有找到匹配节点的儿子节点则产生距离,累加这些距离作为两棵树之间的距离,以此来衡量两个网页之间的相似程度。实验证明本方法可以正确区分同类网页和不同类网页。
英文摘要：
A similarity calculation method for tree-structured web pages is proposed. The structure of web page labels are firstly transformed into tree, and then make the most similar son nodes between each layer of the two trees continue comparing by a dynamic programming algorithm, the nodes which miss match are regarded the part of distance, the total distance between two trees are computed by adding in all the parts of distance through which to calculate their similarity degree. The experimental result shows that this method can effectively and precisely distinguish different web page.
同期刊论文项目
同项目期刊论文
A new query expansion approach using collocation relationships in language models for information re
期刊信息
位置：