基于改进决策树的分布式数据库信息跨域检索研究  

Study on the Cross-domain Retrieval of Distributed Database Information Based on the Improved Decision Tree

作　　者：金黎阳 JIN Liyang(Insurance Center of Daqing Petroleum Management Bureau Co.,Ltd.,Daqing Heilongjiang 163000,China)

机构地区：[1]大庆石油管理局有限公司保险中心,黑龙江大庆163000

出　　处：《信息与电脑》2024年第2期17-19,共3页Information & Computer

摘　　要：针对现有检索方法在对分布式数据库信息跨域检索时,存在实际检索代价过大,需要通过大量节点数量和元组数量的检索才能为用户展示最终检索结果的问题,引入改进决策树,开展对分布式数据库信息跨域检索方法的设计研究。首先将相同语义的检索聚集到同一个检索集合,实现分布式数据库检索聚合;其次利用改进决策树,划分分布式数据库属性;最后基于整理,实现对分布式数据库元数据信息的跨域检索。对比实验结果表明,新的检索方法在实际应用中检索代价与其他现有检索方法相比明显降低,可在检索少量节点和元组后为用户提供其所需检索结果,有效提高检索效率。In view of the existing retrieval method in the distributed database information cross-domain retrieval,the actual retrieval cost is too large,need through a large number of nodes and number of tuple retrieval to show the final retrieval content,introduce the improved decision tree,for the distributed database information cross-domain retrieval method design research.Gather the queries with the same semantics to the same query set to realize the distributed database query aggregation.Using the improved decision tree to divide the distributed database properties.Based on the collation,the crossdomain retrieval of the metadata information of the distributed database is realized.Through comparative experiments,it is proved that the retrieval cost of the new retrieval method is significantly lower compared with other existing retrieval methods.It can provide users with the desired query results after searching a small number of nodes and tuples,and effectively promote the improvement of retrieval efficiency.

关 键 词：改进决策树 数据库 跨域检索 分布式 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]