Abstract: The supervised web-scale forum crawler is to crawl relevant forum content from the web with minimum overhead. Forum threads contain information content that is the target of forum crawlers. each forums have different layouts or styles and have different forum software packages, they always have similar constant navigation paths connected by specific URL types to direct users from entry pages to thread page. we reduce the web forum crawling problem to a URL-type recognition problem. And shows how to learn accurate and effective regular expression patterns of constant navigation paths from automatically created training sets using aggregated results from weak page type classifiers. Robust page type classifiers can be experienced from as few as five annotated forums and applied to a large set of unseen forums.