MoMPCA: MoMPCA: Greedy clustering of count data through a mixture of...

In nicolasJouvin/MMPCA: Inference and Clustering for Mixture of Multinomial Principal Component Analysis

Description

The MMPCA package implements the branch & bound classification-Variational Expectation Maximisation algorithm described in Jouvin et. al. https://arxiv.org/abs/1909.00721. It enables the clustering of counts data such as document/term matrix modeled as a mixture of multinomial PCA model. Model selection is performed via an approximated form of the Integrated Classification Likelihood. .

Details

The main entry point is the mmpca_clust() function to perfom the clustering.

nicolasJouvin/MMPCA documentation built on Jan. 23, 2021, 3 a.m.