Network Camouflage Bad Information Detection Method of Research and Simulation
Qianyi Xu · Jisuanji fangzhen · 2012
Research information detection of bad words in networks and improve the testing accuracy.Traditional bad information detection methods contrast the specific illegal Chinese characters with illegal information Chinese character library,and does not considered the semantic features of Chinese characters.When bad information expressed in legal Chinese characters,the miss detection rate is very high.Thfe paper proposed an information filtering technology based on the semantic association decision.Through calculation the association degree of information semantic and bad information semantic,the semantic factors were used to determine illegal information,which overcomes the disadvantages of traditional methods.The experimental results prove that the method can quickly find out disguised bad information,and ensure the security of information.