RobustEM: Robust Mixture Modeling Fitted via Spatial-EM Algorithm for Model-Based Clustering and Outlier Detection

Share:

The Spatial-EM is a new robust EM algorithm for the finite mixture learning procedures. The algorithm utilizes median- based location and rank-based scatter estimators to replace sample mean and sample covariance matrix in each M step, hence enhancing stability and robustness of the algorithm. To understand more about this algorithm, read the article ''Yu, K., Dang, X., Bart Jr, H. and Chen, Y. (2015). Robust Model- based Learning via Spatial-EM Algorithm. IEEE Transactions on Knowledge and Data Engineering, 27(6), 1670-1682. doi:10.1109/TKDE.2014.2373355''.

Author
Aishat Aloba, Kai Yu, Xin Dang, Yixin Chen, and Henry Bart Jr.
Date of publication
2015-06-05 01:11:35
Maintainer
Aishat Aloba <adetokaloba@gmail.com>
License
GPL-3
Version
1.0

View on CRAN

Man pages

cluster_em
Clustering Algorithm
cluster_em_outlier
Clustering and Outlier Detection Algorithm
cluster_plot
Clustering Plot
confusionMatrix
Confusion Matrix
errorRate
This function calculates the error rates for the different EM...
matchCluster
Correspondence between clusters
normalizedMI
Computes the normalized mutual information
trimmed_bic
Determines the number of components in the distribution.

Files in this package

RobustEM
RobustEM/src
RobustEM/src/wtdr_f.f
RobustEM/NAMESPACE
RobustEM/R
RobustEM/R/trimmed_bic.R
RobustEM/R/em_kotz.R
RobustEM/R/cluster_em_outlier.R
RobustEM/R/k_means.R
RobustEM/R/cluster_em.R
RobustEM/R/em_reg.R
RobustEM/R/conf_mat.R
RobustEM/R/em_rcm.R
RobustEM/R/clusterPlot.R
RobustEM/MD5
RobustEM/DESCRIPTION
RobustEM/man
RobustEM/man/matchCluster.Rd
RobustEM/man/cluster_em.Rd
RobustEM/man/normalizedMI.Rd
RobustEM/man/trimmed_bic.Rd
RobustEM/man/confusionMatrix.Rd
RobustEM/man/cluster_em_outlier.Rd
RobustEM/man/cluster_plot.Rd
RobustEM/man/errorRate.Rd