基于关联规则挖掘的中文文本自动分类

随着电子出版物和互联网文档的飞速增加，自动文档分类工作正变得日渐重要．提出一种基于关联规则的中文文本自动分类方法．该算法将文档视作事务．关键词视作项，利用改进的关联规则挖掘算法挖掘项和类剐间的相关关系．挖掘出的规则形成分类器，可用于类标号未知的文档的区分．实验证明，该算法能较快地获得可理解的规则并且具有较好的召回率和准确率．

Association Rules Based Automatic Chinese Text Categorization

With the rapid expansion of electronic publication, it becomes more and more important to classify document automatically. This paper introduced a new method that is called association based document classification into Chinese Text Categorization. In our algorithms, each document is viewed as a transaction and each keyword as an item, then an association rule mining algorithm is used to mine the correlation between item and category at the end, unlabeled documents are classified using these found rules. Experiments confirmed that this method gets understandable rules of classifier fast and has a promising recall and precision rate.