Abstract: It has spread recently lots of delusional messages, which was the main cause of many Cyber Crimes. Spamming is considered one of the most famous between these crime methods. Besides spamming is annoying, it may be a way to fishing and circumvention. Despite of email providers like Gmail and Hotmail having a lot of techniques for limiting this kind of mails, still there is spam mails found because of circumvention skills of these mails makers, on the detection techniques by changing text shape and content. This paper proposes a Spamming Detection Model – SPM, which is built on semantic text recognition. This model proposes a set of steps to build a new rules and dataset that detect any spamming. In the case of any new spamming, its characteristics will be added as a new data in the special database. The texts will be examined continuously and compared with the semantic rules, which will be updated constantly. It is expected that this proposed model will limit spamming signific