基于无推理机制的集成图查询重写算法

QUERY REWRITING ALGORITHM OF INTEGRATED GRAPH BASED ON NON-REASONING MECHANISM

  • 摘要: 为了实现无监督可视化数据查询,提出一种基于无推理机制的集成图查询重写算法。图的可扩展性能够用视图定义扩展传统的数据集成框架,另外将所有必需的元数据包括全局模式、源描述、映射和查询编码到单个数据结构中,简化了它们之间的互操作性,使得重写算法能够查询此类元数据结构,从而能够有效地识别相关源。实验结果表明所提出的方法具有较好的查询性能,并且不会导致较大的计算成本。

     

    Abstract: In order to realize the unsupervised visualization data query, a query rewriting algorithm of integrated graph based on non-reasoning mechanism is proposed. The extensibility of graph could extend the traditional data integration framework with view definition. In addition, all necessary metadata, including global schema, source description, mapping and query, were encoded into a single data structure, which simplified the interoperability between them, so that the rewriting algorithm could query such metadata structure, and can effectively identify relevant sources. The experimental results show that the proposed method has good query performance and will not lead to large computational cost.

     

/

返回文章
返回