Robust classification of multivariate time series by imprecise hidden Markov models
Robust classification of multivariate time series by imprecise hidden Markov models
复制标题
DOI：
10.1016/j.ijar.2014.07.005
复制
发表时间：
2015-01-01
影响因子：
3.9
通讯作者：
Cuzzolin, Fabio
中科院分区：
文献类型：
--
作者：
Antonucci, Alessandro;De Rosa, Rocco;Cuzzolin, Fabio
提出了一种新的基于隐马尔可夫模型的时间序列分类方法。这些模型的学习是通过耦合EM算法与不精确的Dirichlet模型。在平稳性限制下，每个模型对应于高斯密度的不精确混合，这将问题减少到静态的、不精确的概率信息的分类。两个分类器，一个基于混合物的期望值，另一个对混合物之间的Bhattacharyya距离，开发。这些描述符的界限相对于参数的不精确量化的计算被减少到，分别，线性和二次优化任务，因此有效地解决。分类是通过扩展的k-最近邻方法区间值数据。分类器是可信的，这意味着可以在输出中返回多个类标签。在计算机视觉基准数据集上的实验表明，这些方法达到了所需的鲁棒性，同时优于其他精确和不精确的方法。(C)2014爱思唯尔公司All rights reserved.
A novel technique to classify time series with imprecise hidden Markov models is presented. The learning of these models is achieved by coupling the EM algorithm with the imprecise Dirichlet model. In the stationarity limit, each model corresponds to an imprecise mixture of Gaussian densities, this reducing the problem to the classification of static, imprecise-probabilistic, information. Two classifiers, one based on the expected value of the mixture, the other on the Bhattacharyya distance between pairs of mixtures, are developed. The computation of the bounds of these descriptors with respect to the imprecise quantification of the parameters is reduced to, respectively, linear and quadratic optimization tasks, and hence efficiently solved. Classification is performed by extending the k-nearest neighbors approach to interval-valued data. The classifiers are credal, meaning that multiple class labels can be returned in the output. Experiments on benchmark datasets for computer vision show that these methods achieve the required robustness whilst outperforming other precise and imprecise methods. (C) 2014 Elsevier Inc. All rights reserved.