QQ客服 官方微博 反馈留言

	A LDA Feature Grouping Method for Subspace Clustering of Text Data
	Cai, Yeshou; Chen, Xiaojun; Peng, Patrick Xiaogang; Huang, Joshua Zhexue
	2014
会议名称	2014 Pacific Asia Workshop on Intelligence and Security Informatics, PAISI 2014
会议地点	Tainan, Taiwan
英文摘要	This paper proposes a feature grouping method for clustering of text data. In this new method, the vector space model is used to represent a set of documents. The LDA algorithm is applied to the text data to generate groups of features as topics. The topics are treated as group features which enable the recently publishedsubspace clustering algorithm FG-k-means to be used to cluster high dimensional text data with two level features, the word level and the group level. In generating the group level features with LDA, an entropy based word filtering method is proposed to remove the words with low probabilities in the word distributionof the corresponding topics. Experiments were conducted on three real-life text data sets to compare the new method with three existing clustering algorithms. The experiment results have shown that the new method improved the clustering performance in comparison with other methods. © 2014 Springer International Publishing.(20 refs)
收录类别	EI
语种	英语
内容类型	会议论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/6035]
专题	深圳先进技术研究院_数字所
作者单位	2014
推荐引用方式 GB/T 7714	Cai, Yeshou,Chen, Xiaojun,Peng, Patrick Xiaogang,et al. A LDA Feature Grouping Method for Subspace Clustering of Text Data[C]. 见:2014 Pacific Asia Workshop on Intelligence and Security Informatics, PAISI 2014. Tainan, Taiwan.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。