Self-Organizing Maps with Built-in Missing Data Imputation
The Self-Organizing Maps with Built-in Missing Data Imputation.
Map data to a supervised or unsupervised SOM
missSOM
Calculate distances between object vectors in a SOM
Plot missSOM object
Summary and print methods for missSOM objects
Provides smooth unit colors for SOMs
SOM-grid related functions
The Self-Organizing Maps with Built-in Missing Data Imputation. Missing values are imputed and regularly updated during the online Kohonen algorithm. Our method can be used for data visualisation, clustering or imputation of missing data. It is an extension of the online algorithm of the 'kohonen' package. The method is described in the article "Self-Organizing Maps for Exploration of Partially Observed Data and Imputation of Missing Values" by S. Rejeb, C. Duveau, T. Rebafka (2022) <arXiv:2202.07963>.