NEWS.md

In DCEM: Clustering Big Data using Expectation Maximization Star (EM*) Algorithm

DCEM 2.0.5

DCEM is published in SoftwareX, https://doi.org/10.1016/j.softx.2021.100944. Use citation("DCEM") to cite the package.

Added the functionality to predict the cluster membership for test data.

Fixed a minor bug in co-variance calculation during maximiztion.

DCEM 2.0.4

Added the option to get data membership (maximum posterior probability) from the output.

DCEM 2.0.3

Added quick start examples and use cases in the vignettes.

DCEM 2.0.2

Added the option to get cluster membership of data directly from the output. Patched the code for K-Means++ based initialization.

DCEM 2.0.1

Bug fix release of the DCEM package.

Bug Fixes

Removed the usage of floor in integer division in CPP code. The usage lead to warnings (on Solaris OS) in the previous version - 2.0.0.

DCEM 2.0.0

This is the fourth major release of the DCEM package.

Major Features

Improves the EM* implementation for even faster execution. The EM* is motivated from the ideas published in the Using data to build a better EM: EM* for big data. Hasan Kurban, Mark Jenne, Mehmet M. Dalkilic (2016) https://doi.org/10.1007/s41060-017-0062-1.

The package now supports both the EM* and the traditional EM algorithm for speed-up comparison. The EM* leverages the max-heap structure to expedite the execution time manifold compared to the conventional EM.

DCEM 1.0.0

This is the third stable release of the DCEM package.

Major Features

Implements the EM* algorithm from the ideas published in the Using data to build a better EM: EM* for big data. Hasan Kurban, Mark Jenne, Mehmet M. Dalkilic (2016) https://doi.org/10.1007/s41060-017-0062-1.

The package now supports the EM algorithm and the improved version EM*. The EM* leverages a heap structure to expedite the execution time manifold compared to the conventional EM.

DCEM 0.0.2

This is the second stable release of the DCEM package.

Major Features

Implements the improved initialization schemes (based on the idea published in Kmeans++: The Advantages of Careful Seeding, David Arthur and Sergei Vassilvitskii, http://ilpubs.stanford.edu:8090/778/1/2006-13.pdf.) to expedite convergence on big datasets.

DCEM 0.0.1

This is the first stable release of the DCEM package.

Major Features

Support clustering of both multivariate and univariate data for finite gaussian mixture models.

DCEM documentation built on Jan. 16, 2022, 1:07 a.m.