Robust Nonnegative Matrix Factorization (rNMF)
An implementation of robust nonnegative matrix factorization (rNMF). The rNMF algorithm decomposes a nonnegative high dimension data matrix into the product of two low rank nonnegative matrices, while detecting and trimming outliers. The main function is rnmf(). The package also includes a visualization tool, see(), that arranges and prints vectorized images.

Generalized Linear Mixed Models via Monte Carlo Likelihood Approximation (glmm)
Approximates the likelihood of a generalized linear mixed model using Monte Carlo likelihood approximation. Then maximizes the likelihood approximation to return maximum likelihood estimates, observed Fisher information, and other model information.

Least-Squares Bilinear Clustering for Three-Way Data (lsbclust)
Functions for performing least-squares bilinear clustering of three-way data. The method uses the bilinear decomposition (or biadditive model) to model two-way matrix slices while clustering over the third way. Up to four different types of clusters are included, one for each term of the bilinear decomposition. In this way, matrices are clustered simultaneously on (a subset of) their overall means, row margins, column margins and row-column interactions. The orthogonality of the bilinear model results in separability of the joint clustering problem into four separate ones. Three of these subproblems are specific k-means problems, while a special algorithm is implemented for the interactions. Plotting methods are provided, including biplots for the low-rank approximations of the interactions.

An Adaptive Multi-Scale Basis for High-Dimensional, Sparse and Unordered Data (treelet)
Treelets provides a novel construction of multi-scale bases that extends wavelets to non-smooth signals. It returns a multi-scale orthonormal basis, where the final computed basis functions are supported on nested clusters in a hierarchical tree. Both the tree and the basis, which are constructed simultaneously, reflect the internal structure of the data.

Analysis of Count Time Series (tscount)
Likelihood-based methods for model fitting and assessment, prediction and intervention analysis of count time series following generalized linear models are provided. Models with the identity and with the logarithmic link function are allowed. The conditional distribution can be Poisson or Negative Binomial.

Cluster Detection with Hypothesis Free Scan Statistic (graphscan)
Multiple scan statistic with variable window for one dimension data and scan statistic based on connected components in 2D or 3D.