摘要
随着商业广告短信、色情短信、骚扰短信等通过手机不断地蔓延,严重地影响了人们日常生活和社会的稳定.因此短信分类已经成为自然语言处理的一个重要领域.分析了近年来垃圾短信内容的发展,提出了一种基于字特征的短信分类方法.实验结果表明,和词特征相比,该方法使有用短信的错判率有了明显的降低.总之,字特征用于短信分类是可行的.
With the spread of the commercial advertising messages,pornographic messages and harassing messages through mobile phone,People′s Daily life and social stability have been seriously influenced.Therefore short message classification has become an important field of the natural language processing.The development of short messages content in recent years is analyzed,and a method for short message classification is proposed.The experiment results indicate that in contrast to word features this method reduced the mistake rate of useful message.This means that the method presented in this paper is feasible.
出处
《太原师范学院学报(自然科学版)》
2011年第1期103-105,共3页
Journal of Taiyuan Normal University:Natural Science Edition
关键词
短信分类
字特征
KNN方法
short message categorization
the features of chinese characters
KNN