Extractor:支持查询重构的高效数据库关键词检索系统  

Extractor:A Query-Reformulation Embedded Efficient Keyword Search System over Relational Databases

作　　者：王新军[1] 闫实[1] 彭朝晖[1] 李庆忠[1] 

机构地区：[1]山东大学计算机科学与技术学院,山东济南250101

出　　处：《电子学报》2014年第2期209-216,共8页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61170052;No.61100167);国家科技支撑计划(No.2012BAH54F04);山东省自然科学基金(No.ZR2010FM033)

摘　　要：数据库关键词检索由于具有简便易用的特点,成为数据处理中的一项关键技术和研究热点.目前已有的技术还存在着时间复杂度高、检索结果不够精准等问题.针对存在的问题,本文建立了基于关系数据库的关键词检索系统Extractor.Extractor改进了传统的数据图结构,提出了新的top-k结果树生成和排序机制,提出了基于词关系的查询重构方法.实验证明Extrator具有较好的检索效果和较高检索效率.Keyword Search over Relational Databases （KSORD） ,a more straightforward and user-friendly way of retriving data in relational databases ,has raised much attention and been widely studied in recent years .However ,there are still some prob-lems in the existing techniques .To solve these problems ,Extractor ,a new KSORD system based on datagraph ,is proposed in this paper .The datagraph used by Extractor is derived from exiting datagraphs ,which has been improved a lot and the new features are the bases of the new generating and ranking mechanism of result trees used in Extracor .The interactive query-reformulation method based on word co-occurence is also embedded in Extractor to help users find the right query .Experiments verify the effectiveness and efficiency of Extractor .

关 键 词：关键词检索 关系数据库 查询重构 top-k检索 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]