User alias authority ranking based on email communication relation analysis

Finding information about real identity of email users from emails is one of the important topics in email mining. Existing methods focus extracting name information of users from email datasets, but not pay much attention to the research about aliases that belong to the same user. That extracting user's aliases and ranking authority of the aliases is very significant to some network researches and applications such as identity identifying, important people finding and so on. This paper focuses on ranking the authority of user's aliases having been extracted from emails and gives the most authoritative aliases of users. Based on the facts that different aliases of a user take different positions in the user's email communicating relationship, we propose a novel method to rank the authority of an alias for the email user by analyzing email communication relationship. Results on the public subset of the Enron corpus indicate that the approaches presented in this paper can efficiently rank the authority of user's aliases and find the authoritative aliases of a user and the average precision of the methods is more than 90%.