一种基于XQuery的优化Web信息抽取方法

由于缺乏对页面特征适应性的分析，现有的典型系统难以保障抽取规则的健壮性。提出一种优化的Web信息抽取方法，该方法引入了相互关联的三层规则，在分析页面特征适应性的基础上，从准确率和召回率两方面出发提出了抽取规则的优化算法，并用标准XQuery表达复杂对象抽取规则。实验证明，该方法有效地增强了抽取规则的健壮性及可用性。

Optimized Web information extraction based on XQuery

Due to lack of the analysis of the adaptability of the Web page's characteristics, the current typical systems can hardly provide robust extraction rules. This paper proposed an optimized Web information extraction method which divided rules into three associated layers, suggested an optimized algorithm for extraction rules from the view of the precision and recall ratio through analyzing the adaptability of the page's characteristics, and expressed the complicated object rule in standard XQuery. Experiments indicate that our approach enhances the robustness and usability of the rules.