wanghongjun
Research Associate
|
Supervisor of Master's Candidates
- Master Tutor
- Education Level:PhD graduate
- Degree:Doctor of engineering
- Business Address:犀浦3号教学楼31529
- Professional Title:Research Associate
- Alma Mater:四川大学
- Supervisor of Master's Candidates
- School/Department:计算机与人工智能学院
- Discipline:Electronic Information
Software Engineering
Computer Application Technology
Contact Information
- PostalAddress：
- Email：
- Paper Publications
基于贝叶斯网络的半监督聚类集成模型
- Impact Factor:3.993
- DOI number:10.3724/SP.J.1001.2010.03683
- Affiliation of Author(s):西南交通大学
- Journal:软件学报
- Key Words:半监督聚类集成；变分推理；必连；不连
- Abstract:已有的聚类集算法基本上都是非监督聚类集成算法,这样不能利用已知信息,使得聚类集成的准确性,鲁棒性和稳定性降低.把半监督学习和聚类集成结合起来,设计半监督聚类集成模型来克服这些缺点.主要工作包括:第一,设计了基于贝叶斯网络的半监督聚类集成(semi-supervised cluster ensemble,简称SCE)模型,并对模型用变分法进行了推理求解;第二,在此基础上,给出了EM(expectation maximization)框架下的具体算法;第三,从UCI(University of California,Irvine)机器学习库中选取部分数据来做实验.实验结果表明,SCE模型本身及其变分推理后所设计的EM算法都能进行半监督聚类集成,总的来说,效果比NMFS(algorithm of nonnegative-matrix-factorization based semi-supervised),半监督SVM(support vector machine),LVCE(latentvariable model for cluster ensemble)等算法要好.该半监督聚类集成模型聚集了半监督学习和聚类集成两者的优点,最后的聚类结果比单纯的半监督聚类或聚类集成的效果都要好.
- Co-author:李志蜀,Qi jianhuai,成镲,周鹏,周维
- First Author:wanghongjun
- Indexed by:Academic papers
- Document Code:4075614
- Discipline:Engineering
- First-Level Discipline:Computer Science and Technology
- Volume:21
- Issue:11
- Page Number:2814-2825
- ISSN No.:1000-9825
- Translation or Not:no
- CN No.:11-2560/TP
- Date of Publication:2010-11-15