使用Nokogiri或xpath在没有id或类的网页上解析表格

I wish to parse through a epinions.com page to gather some statistics about a few companies. Epinions have almost no id's or classes, so it's quite difficult to parse the site.I wish to parse through a epinions.com page to