一种改进混合文本密度的网页信息提取方法  

A Web Page Information Extraction Method with Improved Mixed Text Density

作　　者：陈壮 葛斌[1] CHEN Zhuang;GE Bin(School of Computer Science and Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China)

机构地区：[1]安徽理工大学计算机科学与工程学院,安徽淮南232001

出　　处：《佳木斯大学学报（自然科学版）》2022年第1期41-44,153,共5页Journal of Jiamusi University：Natural Science Edition

基　　金：国家自然科学基金项目(51874003);安徽省自然科学基金项目(1808085MG221)资助。

摘　　要：针对当前网页越来越复杂化的特点,提出了一种改进混合文本密度的网页信息提取方法。将Web网页解析成文档对象模型,采用视觉特征去噪的方法对网页进行预处理,通过行列分割的方式对网页进行分块,根据各块文本内容的文本密度、链接密度、链接文本密度、标点符号密度计算出混合密度,通过阈值判断提取出网页中含信息量高的内容信息。该方法能完整提取出网页内容信息,简单高效。实验结果表明,该方法相比于基于统计的CETR提取方法,平均准确率提高了3.68%。In view of the increasingly complex characteristics of current web pages,a web information extraction method with improved mixed text density was proposed.Web page into the document object model,USES the visual features of preprocessing,denoising method to page through the ranks of segmentation approach for Web block,according to each piece of text text density,density of link,the link text density,the density of punctuation,calculate the mixture density by threshold value judgment to extract the high information content information in Web pages.This method can extract the content information of web page completely,which is simple and efficient.Experimental results show that the average accuracy of this method is improved by 3.68%compared with the CETR extraction method based on statistics.

关 键 词：文档对象模型 文本密度 视觉特征去噪 行列分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]