Deep Web集成环境下基于经验值的弹性查询松弛方法

A Flexible Query Relaxation Approach Based on Empirical Value Applied in Deep Web Data Integration Environment

Deep Web环境中,用户输入的查询条件过于严格可能会导致查询的失败,通过查询松弛可以提升查询的有效性.为使查询松弛更加科学准确,提出了一种基于经验值的弹性查询松弛方法.首先获取各查询属性的经验值,并确定各属性的重要度;然后给出了基于经验值的弹性查询松弛策略,即按照属性重要度由低到高的顺序,依次在属性的经验值范围内对查询条件进行弹性松弛,使查询松弛后返回的结果尽可能与用户的查询请求相匹配;最后通过实验验证了所提查询松弛方法切实可行.

In the environment of Deep Web,because the query conditions given by the users are so strict that may cause the failure of the query.The effectiveness of the query can be improved by query relaxation.In order to make the query relaxation more scientific and accurate,a method of flexible query relaxation based on the empirical value is proposed in this paper.Firstly,get the empirical value of the attributes,and then determine their important degree.Secondly,a flexibility query relaxation strategy based on the empirical value is given.In this strategy,relax the query requirement according to the importance of attributes from top to bottom in the range of the empirical value of attributes,and then make the result got by the query relaxation match the requirement of the user as possible.Finally,use an experiment to verify the query relaxation method proposed in this paper is feasible.