【摘要】 在面向对等网络知识集成网络的检索过程中,为了实现知识获取效率和语义一致性之间的均衡,提出一种融合分类检索、语义路由和加载推理的检索机制。该机制根据网络拓扑结构的特点定义面向不同范围的检索分类,将检索范围进行适度控制;基于语义匹配得到满足检索需求的知识节点路由列表,对节点加载范围进行选择性限定,以获得更高的检索效率和更合理的检索结果集;建立加载推理机制,在上下文语义模型的支持下,支持复杂检索语义的明确表达。该方法充分利用集成网络的拓扑结构特点和知识集成网络的语义推理特性,在降低搜索时间和空间消耗的基础上,提升了多节点检索结果的语义一致性和语义推理能力。更多还原

【Abstract】 In order to achieve equilibrium between knowledge acquisition efficiency and semantic consistence in query process of Peer-to-Peer(P2P)-oriented knowledge integration network,a query approach combing classified query,semantic routing and semantic inference mechanisms was proposed.Accoring to characteristics of network topological structure,the approach defined query category oriented to different scopes and limited the scope of query.Knowledge node routing lists based on semantic matching which satisfied query requirements were obtained.Semantic scope was limited selectively in order to obtain higher query efficiency and more reasonable query result sets.Semantic inference mechanism was established.With the help of contextual semantic models,it supported formal expression of complicated query semantics.This query mechanism made full use of topological structure characteristics of integrated network and semantic reasoning feature of knowledge integration network,it reduced search time and space consumption,enhanced semantic consistence and semantic reasoning ability.更多还原