Extensible Markov Model for Modelling Temporal Relationships Between Clusters

Implements TRACDS (Temporal Relationships between Clusters for Data Streams), a generalization of Extensible Markov Model (EMM). TRACDS adds a temporal or order model to data stream clustering by superimposing a dynamically adapting Markov Chain. Also provides an implementation of EMM (TRACDS on top of tNN data stream clustering). Development of this package was supported in part by NSF IIS-0948893 and R21HG005912 from the National Human Genome Research Institute.


Changes in version 1.0-11 (07/23/2015) * Fixed NAMESPACE (non-standard imports)

Changes in version 1.0-9 (5/15/2015) * Fixed dependencies in NAMESPACE * Fixed problems with testthat

Changes in version 1.0-8 (1/14/2014) * score: match_clusters can now be "exact", "nn" or "weighted", where weighted replaces all the weighted scores. * added smooth_transitions (experimental) * plus_one is now called prior (since it adds a uniform prior for transition probabilities) * score has now an argument called normalize

Changes in version 1.0-7 * improved plot for tNN * tNN: centroids stop moving to prevent 2 clusters from colliding * fixed bug with getting transitions from a TRACDS object with 0 transitions * find_clusters also accepts now a threshold factor for match_clusters * score: weights for scoring are now normalized for threshold * transition from igraph0 to igraph

Changes in version 1.0-6 * Improved performance for prune() * Added object.size() which also reports memory used by the model * Switched to igraph0

Changes in version 1.0-5 * Added several methods for score. * Threshold for pruning and finding rare clusters/transitions is now less or equal instead of just less

Changes in version 1.0-3 * Finished the transition to reference objects implemented using environments. * transition_table() now has for consistency also the default of add_one set to FALSE. * last_clustering() was added as an accessor function to tNN. * 'igraph' is now the default plot method. * Added combining EMMs. * Added reclustering using tNN. * Added coloring of states and transitions to plot(). * Added coercion functions as.igraph() and as.graph from TRACDS.

Changes in version 1.0-2 * service release

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.0-11 by Michael Hahsler, 5 years ago


Browse source code at https://github.com/cran/rEMM

Authors: Michael Hahsler [aut, cre, cph] , Margaret H. Dunham [aut, cph]

Documentation:   PDF Manual  

GPL-2 license

Imports methods, stats, cluster, clusterGeneration, MASS, utils

Depends on proxy, igraph

Suggests graph, Rgraphviz, testthat

See at CRAN