Nowadays the amount of spam is increasing in an uncontrollable way. This situation decreases the email trust of the Internet users and reduces the usability of the emails to the critical level. On the other hand, this makes a lot of money loss for the hosting companies. There are many anti-spam tools that are developed against spammers. Some of these tools are really succesful. However; since the spammers improve their techniques, spams gain immune to these tools. In this thesis, the problem of detecting spams in in-coming and out-going emails is adressed.. To achieve this goal, an email social network is constructed by using the traffic of emails between users.While constructing this network, only information that can be gained from the email structure is used. The results on the real data set show that the techniques applied have been effective and also point to new directions of research in this area.