Research of Spam Filtering Based on Machine Learning
Email has become an essential means of communication . However,the spam has reduced the effectiveness of email as a communications medium. Recently,spam filters have been widely adopted as a means of combating these unwanted messages. This paper presents a model for better spam detection by combining the classical naive Bayesian filter with a neural network that analyzes various characteristics of the email subject. Neural network design and naive Bayesian classifier design have been described in more detail Practice has proved that this model is better than the traditional neural networks and the naive Bayesian method,because this model uses a complex method and a multi-method multi-layer filter. It will be an effective method of spam filtering and the future direction.
spam filtering naive Bayes neural network
Ren Jiansi
School of Computer,China University of Geosciences,Wuhan 430074,China
国际会议
南宁
英文
255-257
2010-12-10（万方平台首次上网日期，不代表论文的发表时间）