中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
A comparative study on clustering-based classification algorithms

文献类型:期刊论文

作者Sun zhaocai; Liu zhi; Ye yunming; Deng shengchun; Huang zhexue
刊名ICIC Express Letters
出版日期2011
卷号5期号:11页码:3987-3994
英文摘要For most classifiers, they are based on underlying assumptions or models. If the model or assumption matches the true data distribution, the accuracy is high, and vice versa. That is the problem of model-misfit. This paper uses clustering or cluster analysis in classification to solve the problem. For this reason, we propose a general clustering-based classification framework. In the framework, clustering algorithm is used to re-collect the data, like a filter. Thus, the diversities between classes are weakened. By that, Model-misfit is no problem any more. In our framework, almost all clustering and classification algorithms can be integrated together for the better performance. In this paper, we present an empirical study on four clustering-based classification methods. On complex data (e.g., non-linear), experimental results show that the clustering-based approach can improve the performance of the traditional classifier, especially for simple classifiers (e.g., k-NN). © 2011 ICIC International.(9 refs)
收录类别EI
语种英语
源URL[http://ir.siat.ac.cn:8080/handle/172644/3552]  
专题深圳先进技术研究院_数字所
作者单位ICIC Express Letters
推荐引用方式
GB/T 7714
Sun zhaocai,Liu zhi,Ye yunming,et al. A comparative study on clustering-based classification algorithms[J]. ICIC Express Letters,2011,5(11):3987-3994.
APA Sun zhaocai,Liu zhi,Ye yunming,Deng shengchun,&Huang zhexue.(2011).A comparative study on clustering-based classification algorithms.ICIC Express Letters,5(11),3987-3994.
MLA Sun zhaocai,et al."A comparative study on clustering-based classification algorithms".ICIC Express Letters 5.11(2011):3987-3994.

入库方式: OAI收割

来源:深圳先进技术研究院

浏览0
下载0
收藏0
其他版本

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。