摘要
目前XML查询语言及查询界面对Web用户过于复杂,该文描述了一种XML文档索引机制,在此基础上建立了一个通用的贝叶斯网络查询模型。用户只需在交互界面输入自然语言描述的查询,系统就能对其实现基于语义的构造,由它生成多个结构化查询;对这些查询建立贝叶斯网络,计算查询在给定文档下的概率,选择概率最大的前3个查询提交给系统执行。
Recently, the query languages and interfaces for XML are too intricate for Web users. This paper presents an index mechanism for XML documents, builds a retrieval model based on Bayesian network. The model constructs the user＇s query based on semantic in natural language, generates several structured queries, and makes Bayesian network for queries, and computes the probabilities on the document collection. Then the system selects the top there queries to execute.
出处
《计算机工程》
EI
CAS
CSCD
北大核心
2006年第15期67-69,共3页
Computer Engineering
基金
河北省科学技术研究与发展基金资助项目(04213534)
