一种基于《知网》的中文文本聚类算法的研究

Research of novel Chinese text clustering algorithm based on HowNet

针对基于关键词集的中文文本聚类算法中存在的问题,将《知网》引入到中文文本的特征表示中,并在此基础上提出了一种基于《知网》的中文文本聚类算法。该算法在中文文本表示中加入了基于《知网》的概念特征,实验结果表明该算法能够更好地将语义相关的中文文档聚集在一起,与传统的基于关键词集的中文文本聚类算法相比,聚类质量得到了较大提高。

To settle the problem of Chinese text clustering algorithm based on keywords set,this paper introduces HowNet into the representation of Chinese text representation and presentes a Chinese text clustering algorithm based on HowNet.This algorithm adds the conceptual characteristic based on Hownet to the representation of Chinese text.Experimental results show that this algorithm can cluster the semantic relative Chinese text into the same cluster better and improve the quality of text clustering greatly.