Changes in version 2.0.6 (2026-04-14)                  

Bug Fixes

  - Reformatted NEWS.md to comply with CommonMark and CRAN requirements.
  - Incremented version to resolve archival status.
  - Fixed CITATION file entries to use modern bibentry() syntax.
  - Removed non-standard files from package build (.github,
    CRAN-RELEASE, cran-comments.md).

                 Changes in version 2.0.5 (2022-01-15)                  

  - 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 maximization.

                 Changes in version 2.0.4 (2020-08-02)                  

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

                        Changes in version 2.0.3                        

  - Added quick start examples and use cases in the vignettes.

                 Changes in version 2.0.2 (2020-04-07)                  

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

                 Changes in version 2.0.1 (2020-02-28)                  

Bug Fixes

  - Removed the usage of floor in integer division in CPP code to
    resolve Solaris OS warnings.

                 Changes in version 2.0.0 (2019-11-29)                  

  - This is the fourth major release of the DCEM package.
  - Improves the EM* implementation for faster execution based on Kurban
    et al. (2016) https://doi.org/10.1007/s41060-017-0062-1.
  - Supports both EM* and traditional EM algorithm for speed-up
    comparison.

                 Changes in version 1.0.0 (2019-07-24)                  

  - Implements the EM* algorithm leveraging a heap structure to expedite
    execution time.

                 Changes in version 0.0.2 (2019-04-05)                  

  - Implements improved initialization schemes based on K-means++
    (Arthur and Vassilvitskii).

                 Changes in version 0.0.1 (2018-09-30)                  

  - Initial stable release.
  - Support for clustering multivariate and univariate data for finite
    Gaussian mixture models.