利用同领域数据源主题更新的关联特点,提出了一种非合作结构化深网数据源摘要的动态更新方法,在保证数据源选择效果的前提下,较大幅度地提高了数据源摘要更新的效率(即减少了数据源摘要更新的工作量).实验结果表明,该方法可以减少87.7%以上的摘要更新工作量,同时具有较好的召回率及准确率.
We propose a dynamic update method of summary for non-cooperative structured deep web selection with the associated update characteristics of data sources in a field. Our method significantly improves the efficiency of the summary update under the premise of ensuring the effect of the data source selecting. The experiment results show that our dynamic update method of summary reduce the calculation work by 87.7% and has a good recall ratio and precision.