基于中心抽样的KNN算法在文本分类中的应用

The Application of KNN Algorithm Based on Central Sampling in Text Categorization

文本分类是文本数据挖掘中的一个重要的内容,现阶段文本分类用到的主要算法有KNN,贝叶斯,神经网络等。KNN算法因为原理简单,分类效果较好,在文本分类中得到应用,但在数据量大时其运行效率上存在一定的局限性,本文提出一种基于中心抽样的KNN算法,并用20newsgroup数据集对其进行验证,在不影响准确率的情况下,提高了运行效率,取得了不错的效果。

Text classification is an important part of text data mining.At this stage,the main algorithms used in text categorization include KNN,Bayesian,neural networks.KNN algorithm for its simple ideas,with good efficiency,has an important application in text classification.But the KNN algorithm has certain limitation in the operation efficiency,especially dealing with a large amount of data.In this paper,a KNN algorithm based on central sampling is proposed,which is verified by 20newsgroup data set.It can improve the operation efficiency without affecting the accuracy,and has achieved good results.