Today, the spread of internet technology in developing electronic communication has brought some problems. One of the most important problems of electronic communication on the Internet is to spread unwanted messages known as SPAM. In this study, an expert system is designed for filtering SPAM and its results are given. Firstly, the literature is reviewed related to SPAM filtering softwares. Later, the expert system designed application is explained step by step. A dataset consisting of 4601 electronic mail was prepared including SPAM and normal e-mail. Content of each e-mail in dataset is analyzed and attributes in texts are detected. Developed system decide whether this attributes are SPAM or not by electing from this attributes and by regarding IP address of e-mails. The success of this study is put forth with real time applications.