This paper reviews the stages of development and common approaches to spam text classification. This was followed by an introduction to the basic knowledge of spam filtering. After that, some research results are presented. The last part will be a discussion of the literature review and some suggestions for the future development of spam filtering techniques. This paper found that the future direction of spam filtering should be a combination of machine learning techniques and deep learning techniques and propose to take into consideration diversified features of the emails other than text such as IP addresses and IP reputation values into the machine learning models.