A Novel Approach To Automatically Extracting Main Content of Web News

摘要

Recently, the Web has been the data repository. In order to obtain the relevant information from the repository, many research have been made. The typical function of Web news extraction is to locate the useful content text and filter the noises , both main issues result in Web news extraction that is an open research problem. In this paper , we describe an approach that can cluster the pages which share common extracting path and automatically extract location of main text passages. Our approach can apply to structural Web pages . Moreover, we developed an extracting system by using our algorithm. Experiments are done over several important on-line news sites and experimental results on our extracting system show that the approach can achieve higher extraction accuracy than RTDM algorithm.