The Research Of Algorithm Over Spam Filtering Method

In company with the popularization of Internet, e-mail, has gradually developed into one of the most significant means of communication for people's work and daily life, due to its rapidness and convenience. However, the consequent junk mail problem turns to be austerity increasingly, either. Regarded as the bearer of commercial advertisement,worm or sensitivity visceral, it causes serious threaten to systemic security and personal life as the time past by. Research over Anti-spam tends to be a global significance subject.Support Vector Machine is a newly-developed pattern recognition method based on the statistics learning theory. It reflects peculiar advantages when solving finite examples, non-linearity and high-dimensional pattern recognition issues. It takes into account the requirement for extension capability while pursues the most optimal outcome under the condition of finite examples. Considering the sort performance strongly depends on the parameters chosen for kernel function, genetic algorithm is used to optimize the kernel function's parameters of SVM. The optimize algorithm is distributed in the junk mail filter model. The purpose of modeling is for research only, and is experimented to observe its applicability and effectiveness. The primary experiment shows that the model performances well and its training time is short.This paper firstly introduces the basic knowledge of spam, including its definition, developing history,imperil, and the existed spam filtering method. SVM based filtering method mainly belongs to the content filtering field, so the text category and machine learning knowledge are expounded. Secondly, SVM theory and pretreatment work of spam examples are also introduced. SVM based spam filtering algorithm is inducted over the SVM theory and the dependence of SVM is analyzed in purpose to redistribute SVM algorithm, using genetic algorithm to optimize the parameter of kernel function. In the implementation part, the outline codes are given. Finally, we evaluate the...