一种基于非参数贝叶斯模型的聚类算法
PDF下载 (454)张媛媛.一种基于非参数贝叶斯模型的聚类算法[J].宁波大学学报(理工版),2013,26(04):24-28.DOI:
ZHANG Yuan-yuan.Data Clustering via Nonparametric Bayesian Models[J].Journal of Ningbo University(Natural Science & Engineering Edition),2013,26(04):24-28.DOI:
| Title: | Data Clustering via Nonparametric Bayesian Models |
| 作者: | 张媛媛 |
| Author(s): | ZHANG Yuan-yuan |
| 关键词: | 非参数贝叶斯模型; Dirichlet过程混合模型; 聚类分析; Gibbs采样 |
| Keywords: | nonparametric Bayesian models; Dirichlet process mixture model; clustering analysis; Gibbs sampling |
| 分类号: | TP391 |
| 文献标识码: | A |
| 摘要: | 鉴于聚类分析是机器学习和数据挖掘领域的一项重要技术, 并且与监督学习不同的是聚类分析中没有类别或标签的指导信息, 所以如何选择合适的聚类个数(即模型选择)一直是聚类分析中的难点. 由此提出了一种基于Dirichlet过程混合模型的聚类算法, 并用collapsed Gibbs采样算法对混合模型的参数进行估计. 新算法基于非参数贝叶斯模型的框架, 能够在不断的采样过程中优化模型参数并形成合适的聚类个数. 在人工合成数据集和真实数据集上的聚类实验结果表明: 基于Dirichlet过程混合模型的聚类算法不但能够自动确定聚类个数, 而且具有较强灵活性和鲁棒性. |
| Abstract: | Clustering is one of the most useful techniques in machine learning and data mining. In cluster analysis, model selection concerning how to determine the number of clusters is an important issue. Unlike supervised learning, there are no class labels and criteria to guide the search, so the model for clustering is always difficult to select. To tackle this problem, we present the concept of nonparametric clustering approach based on Dirichlet process mixture model (DPMM), and apply a collapsed Gibbs sampling technique to sample the posterior distribution. The proposed clustering algorithm follows the Bayesian nonparametric framework and can optimize the number of components and the parameters of the model. The experimental result of clustering shows that this Bayes model has promising properties and robust performance. |
| 参考文献 /References: | [1] Jain K A. Data clustering: 50 years beyond K-means[J]. Pattern Recognition Letters, 2010, 31:651-666. [2] Bishop M C. Pattern recognition and machine learning [M]. New York: Springer Press, 2007. [3] Claeskens G, Hjort N L. Model selection and model averaging[M]. New York: Cambridge University Press, 2008. [4] Shahbaba B, Neal R. Nonlinear models using Dirichlet process mixtures[J]. The Journal of Machine Learning Research, 2009, 10:1829-1850. [5] Fox E, Sudderth E, Jordan M, et al. Nonparametric Bayesian learning of switching linear dynamical systems [C]. In Advances in Neural Information Processing Systems 21, Vancouver and Whistler, BC, Canada: The MIT Press, 2008. [6] Sudderth E. Graphical models for visual object recognition and tracking[D]. MA, Cambridge: Massa- chusetts Institute of Technology, 2006. [7] Gershman S, Blei D. A tutorial on Bayesian nonpara- metric models[J]. Journal of Mathematical Psychology, 2012, 56:1-12. [8] Hjort L N, Holmes M P, Walker G S. Bayesian nonparametric[M]. New York: Cambridge University Press, 2010. [9] Ferguson S T. A Bayesian analysis of some nonparametric problems[J]. The Annals of Statistics, 1973, 12:209-230. [10] 周建英, 王飞跃, 曾大军. 分层Dirichlet过程及其应用综述[J]. 自动化学报, 2011, 37(4):390-407. [11] Teh W Y. Dirichlet processes[C]. Machine Learning Summer School Tutorial and Practical Course, 2007. [12] Teh W Y, Jordan I M, Beal J M, et al. Hierarchical Dirichlet processes[J]. Journal of the American Statistical Association, 2006, 101:1566-1581. [13] Rasmussen E C. The infinite Gaussian mixture model[C]. In Advances in Neural Information Processing Systems 12, MIT Press, 2000:554-560. [14] Teh W Y. Bayesian nonparametric: DP mixtures[EB/OL]. [2009-10-11]. http://www.gatsby.ucl.ac.uk/~ywteh/teaching/ npbayes.html. [15] Neal R. Markov chain sampling methods for Dirichlet process mixture models[J]. Journal of Computational and Graphical Statistics, 2000, 9:249-265. [16] Blei M D, Jordan I M. Variational inference for Dirichlet process mixtures[J]. Bayesian Analysis, 2006(1):121-144. [17] Blei M D, Frazier P. Distance dependent Chinese restaurant processes[J]. The Journal of Machine Learning Research, 2011, 12:2461-2488. [18] Rasmussen, C, Williams C M. Gaussian processes for machine learning[M]. New York: Springer Press, 2006. |
| 备注/Memo: | 收稿日期: 2013-06-09. 基金项目: 国家自然科学基金（61175026）; 浙江省新一代移动互联网用户端软件科技创新团队项目（2010R50009）; 宁波市自然科学基金（2011A610193）; 宁波大学学科项目（XKL09154）. 作者简介: 张媛媛（1981-）, 女, 浙江宁海人, 博士研究生/实验师, 主要研究方向: 机器学习和计算机视觉. E-mail: zhangyuanyuan@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
