摘要
介绍了传统的kNN的文本分类方法,分析其实质,指出其不足,提出了一种基于隐含语义的改进方案,并结合实际给出实验结果。
This paper introduces the traditional kNN text classification and analyses the essential of it.For the shortage of traditional methods,the authors put forward an improved k-NN text classification based on latent semantic.The experi-mental results is given using the new method in the end.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第6期71-73,共3页
Computer Engineering and Applications
基金
国家高性能计算基金资助项目(编号:99319)
关键词
KNN
奇异值分解
文本分类
k Nearest Neighbor,Singular Value Decomposition,Text Classification