摘要
设计并实现中文垃圾短信过滤器,能够较好识别不断变化的垃圾短信。以逻辑回归模型为基础,提出字节级n元文法提取短信特征,并采用TONE(Train On or Near Error)方法训练过滤器。通过实验测试,证明应用该方法实现的垃圾短信过滤效果很好。
We designed and implemented the Chinese short messages service spam filter which can defense the evolution of short message spam.Logistic regression model is used as its filtering model;byte level N-gram is put forward to extract message spam's features;and the filter is trained with TONE(Train On or Near Error) method.The performance evaluated by experiments of the short messages spam filter based on the method is pretty good.
出处
《黑龙江工程学院学报》
CAS
2010年第4期36-39,共4页
Journal of Heilongjiang Institute of Technology
基金
黑龙江省教育厅资助项目(11551401)
关键词
中文垃圾短信过滤
逻辑回归模型
N元文法
TONE
Chinese short messages service spam filtering
logistic regression model
N-gram
TONE