In 2003, et al. Jérôme R. Bellegarda, demonstrated conventional mail filtering techniques based on unsupervised learning where classification is done based on keyword matching. But if spammers change spam framing tricks, old classifiers will no longer be able to give accurate results. This is the worst part of unsupervised learning. On the other hand, in the same paper, machine learning techniques based on supervised learning are introduced where the classifiers are regularly fed the changing spam patterns with different datasets[15].In 2006, et al . Giorgio Fumera, focused in his work in [20] on text categorization techniques based on machine learning and pattern recognition approaches for analyzing the semantic content of emails instead of coded rules manually derived from spam email analysis. This article shed light on the concept of content-based spam filtering and spam filtering that leverages textual information embedded in images sent as attachments. In 2009, et al. Ronald Bhuleskar trumpeted a new approach to the HSF model. This is a combinatorial filter model of various spam filtering techniques. The author used unsupervised and supervised techniques simultaneously in his model. It filters an incoming mail through different filters separately, but all filters must be arranged in parallel. The parallel filters used in this article were Black and White List, Content-Based Filtering, and Forging Filtering. During falsification, the sender's IP address is verified and then at the server level, validation of the domain name of the email sending server with its IP address or reverse DNS lookup[18]. In 2010, et al. Morteza Zi Hayat showed it in [19], here again supervised learning is used and promoted. In this environment...... middle of article ......s and Networks, IEEE Computer Society, 2009, pp. 302-307.[19] Morteza Zi Hayat, Javad Basiri, Leila Seyedhossein, Azadeh Shakery, "Content-based concept drift detection for email spam filtering", 5th International Telecommunications Symposium (IST'2010), 2010, pp. 531-536.[20] Giorgio Fumera, Ignazio Pillai, Fabio Roli, “Spam filtering based on the analysis of textual information embedded in images,” in Journal of Machine Learning Research, Vol. 7, December 2006, p. 2699-2720.[21] Zhenyu Zhong, Kang Li, “Accelerate statistical spam filter by approximation”, i.e. transactions on computers, vol. 60, no. January 1, 2011, p. 120-133.[22] Basheer Al-Duwairi, Ismail Khater, Omar Al-Jarrah, “Image spam detection using image texture features”, International Journal for Information Security Research (IJISR), Volume 2, Issues 3/4 , September/December 2012, pp.. 344-353.