Design of Parallel Filtering Model of Spam
Chun Liu · Journal of Daqing Normal University · 2008
With the growing popularity of the network,the problem of spam is increasingly serious and corresponding various spam filtering algorithm come into being.Naive Bayes method is simple but limited because of algorithm assumptions,so its recall rate and the correct rate difficult to achieve a higher level.Other algorithms are more effective but complicated to compute.Trying to resolve this problem,we propose using slipping window,just like the pipeline of computer theory,and we call it as parallel filtering model.Finally,we get satisfactory results.