网页内容安全快速信息抽取方法

文章提出一种基于静态网页特征的文本信息抽取方法。该方法首先根据静态网页的URL特征判断其是否是静态网页，然后根据静态网页的结构特征和内容特征对标题和正文文本内容进行抽取．再按照统一规范将结果顺序存储便于再处理。实验结果表明，网页内容信息抽取的查全率和查准率分别为96．2％和95．9％，该方法计算量小、抽取速度快、正确率高，可实际应用于大规模的网页内容安全分析。

Research on Rapid Information Extractionin Web Content Security

This paper proposes a new text information extraction algorithm based on the characteristics of static web page.This method can boost the efficiency of static web page recognition and text information extraction according to the URL,structure and content features,and then it will store them sequentially according tothe standard formats.The experimental results show that the algorithm can extract web text information perfectly with the recall and precision ratioreaching up to 96.2% and 95.9%.This method has a small amount of computation and works fast and accurately,which can be applied directly to the large-scale analysis of web contentsecurity.