benchmarks-functions

benchmark.random

benchmark.kmeans_lda

an S4 object of class <code><a href="/link/?package=MoMPCA&version=1.0.1" mmpcaclust'="" data-mini-rdoc="MoMPCA::"></a>'>mmpcaClust</code>

Some argument to be consistent with the function's skeleton :
<code>K</code> and <code>nruns</code> are optional arguments for some of them.

Number of topics (dimension of the latent space).

Number of restart of the <code><a rd-options="stats" href="/link/kmeans?package=MoMPCA&version=1.0.1&to=stats" data-mini-rdoc="stats::kmeans">kmeans</a></code>()
algorithm.

nruns

These are wrapper to other methods for the clustering of count
 data. They can be used to initialize the clustering. It is also possible to
 implement your own benchmark function depending on other packages.

Cluster any count data matrix with a fixed number of variables, such as document/term matrices. It integrates the dimension reduction aspect of topic models in the mixture models framework. Inference is done by means of a greedy Classification Variational Expectation Maximisation (C-VEM) algorithm. An Integrated Classication Likelihood (ICL) model selection is designed for selecting the latent dimension (number of topics) and the number of clusters. For more details, see the article of Jouvin et. al. (2020) <arxiv:1909.00721>.

Nicolas Jouvin

MoMPCA

Inference and Clustering for Mixture of Multinomial Principal
Component Analysis

benchmarks-functions function

Random initialisation of the clustering. Arguments K and nruns are unused

<code>benchmark.random</code>

Cluster the matrix theta obtained by a
 topicmodels LDA with K topics

<code>benchmarks.kmeans_lda</code>

an S4 object of class <code><a href='<a href='mmpcaClust'></a>'>mmpcaClust</a></code>

Number of restart of the <code><a rd-options='stats' href='kmeans'>kmeans</a></code>()
algorithm.

benchmarks-functions: Benchmarks functions for clustering

Description

Usage

Arguments

Value

<code>benchmark.random</code>

<code>benchmarks.kmeans_lda</code>