Detecting Near Duplicates for Web Crawling － simhash与重复信息识别