科研成果详情

题名A hierarchical infinite generalized dirichlet mixture model with feature selection
作者
发表日期2014
会议名称3rd International Conference on Adaptive and Intelligent Systems (ICAIS)
会议录名称Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN0302-9743
卷号8779 LNAI
页码1-10
会议日期SEP 08-10, 2014
会议地点Bournemouth, ENGLAND
摘要

We propose a nonparametric Bayesian approach, based on hierarchical Dirichlet processes and generalized Dirichlet distributions, for simultaneous clustering and feature selection. The resulting statistical model is learned within a variational framework that we have developed. The merits of the developed model are shown via extensive simulations and experiments when applied to the challenging problem of images categorization. © 2014 Springer International Publishing Switzerland.

关键词Clustering feature selection generalized Dirichlet image databases mixture model variational inference
DOI10.1007/978-3-319-11298-5_1
URL查看来源
收录类别CPCI-S
语种英语English
WOS研究方向Computer Science
WOS类目Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications
WOS记录号WOS:000346932400001
Scopus入藏号2-s2.0-84958551550
引用统计
被引频次:1[WOS]   [WOS记录]     [WOS相关记录]
文献类型会议论文
条目标识符https://repository.uic.edu.cn/handle/39GCC9TT/13287
专题个人在本单位外知识产出
理工科技学院
作者单位
1.Department of Computer Science and Technology,Huaqiao University,China
2.Al Imam Mohammad Ibn Saud Islamic University (IMSIU),Riyadh,Saudi Arabia
3.Concordia University,Montreal QC,Canada
4.Taif University,Taif,Saudi Arabia
推荐引用方式
GB/T 7714
Fan, Wentao,Sallay, Hassen,Bouguila, Nizaret al. A hierarchical infinite generalized dirichlet mixture model with feature selection[C], 2014: 1-10.
条目包含的文件
条目无相关文件。
个性服务
查看访问统计
谷歌学术
谷歌学术中相似的文章
[Fan, Wentao]的文章
[Sallay, Hassen]的文章
[Bouguila, Nizar]的文章
百度学术
百度学术中相似的文章
[Fan, Wentao]的文章
[Sallay, Hassen]的文章
[Bouguila, Nizar]的文章
必应学术
必应学术中相似的文章
[Fan, Wentao]的文章
[Sallay, Hassen]的文章
[Bouguila, Nizar]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。