摘要
文中介绍了一种用SVM进行主动学习的方法 ,解决在某些机器学习问题中 ,训练样本获取代价过大带来的问题。实验表明 ,该方法与普通SVM方法相比 ,在保证SVM分类器性能的前提下 ,可有效减少学习所需的样本数量。最后设计了一个基于该思想的邮件过滤器模型 ,依据该模型设计的邮件过滤器将有实时监控、自动更新邮件过滤模块的能力。
An active learning method using SVM is introduced in this paper. It is used to solve the problem of the excessive expenses caused by obtaining the examples in the machine learning. Experiment shows,compared with the general SVM,it can reduce the number of examples effectively on the premise of keeping correctness of the classifier. Finally,an email filter model based on this idea is brought forward. An email filter designed according to the model is capable of real-time monitoring and renewing the filter module automatically.
出处
《计算机应用》
CSCD
北大核心
2004年第1期1-3,共3页
journal of Computer Applications
关键词
支持向量机
主动学习
文本分类
邮件过滤
support vector machine
active learning
text classification
email filter
