Liang Li, Yan Jianwei, Nie Ying
(School of Electronics and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China)
Abstract: A method for mining junk e-mails based on association rules was
proposed. The supporting rate of key words and source addresses were computed for
determining junk e-mails. Adding restraints of source addresses and key words, the
algorithm is improved on the basis of Apriori algorithm. The algorithm is more accurate
than the e-mails filtering algorithms based on key words, and its complexity is less than
the algorithm based on semantic analysis. The experiments results show that the error
judgment rate is reduced to 4 percent when the number of e-mails reaches to 350, and the
filtering speed is increased with increasing the amount of e-mails.
Keywords: junk e-mail; source address; restraint; filtering