Make a summary database of nightclubs, parsed from 2 webdire

There are 2 separate directories of nightclubs and bars. Need to retrieve all clubs from both websites. Doubling records, which describe the one club have to be joined. Writings of club names can differ, that's why need to find doubles by comparing names, addresses, phone numbers, websites. Provide final result in mysql dump format and XML (the last one - for human reading, XML standart which you will use doesn't matter).