Web search engines return lists of web pages sorted by the page’s relevance to the user query. The problem with web search relevance ranking is to estimate relevance of a page to a query. Nowadays, commercial web-page search engines combine hundreds of features to estimate relevance. The specific features and their mode of combination are kept secret to fight spammers and competitors. Nevertheless, the main types of features at use, as well as the methods for their combination, are publicly known and are the subject of scientific investigation.