index.md

In jansodoge/awesom_dev_version: Interactive Self-Organizing Maps

aweSOM

aweSOM is an R package offering a set of tools to explore and analyze datasets with Self-Organizing Maps (also known as Kohonen maps), a form of artificial neural network originally created by Teuvo Kohonen in the 1980s. The package introduces interactive plots, making analysis of the SOM easier.

aweSOM provides a variety of functions to analyze and visualize SOMs.

• Initializing SOMs on training data and training these using the kohonen package.
• Interactive visualizations of structural characteristics and distribution of numeric as well as categorical variables of SOMs
• Exporting interactive SOM visualizations as standalone HTML file, .png or .svg.
• Performing and graphically evaluating super-clustering on SOM using hierarchical and PAM clustering methods
• Ability of generating replicable R-code using the web-based interface

These can be used either through the web-based interface (called by aweSOM()) or through command-line functions. The package relies on the kohonen package for the training of SOMs.

Install

Install the developer version of aweSOM from Github using the devtools package

devtools::install_github("jansodoge/awesom_dev_version")

aweSOM web-based interface

An intuitive and user-friendly approach to training and analyzing datasets with self-organizing maps is enabled by the aweSOM web-based interface. It allows the user to import data, train a SOM, analyze via the interactive visualizations and download results. Plots can be exported as interactive versions (HTML format), or static image files (.png or.svg). Additionally, it creates replicable R-code to perform the operations using command-line functions of aweSOM.

The web-based interface can be launched using the following function.

aweSOM::aweSOM()

aweSOM command-line functions

The aweSOM package provides command-line functions that enable functionality similar to the web-based interface. Detailed documentation of the command-line functions and their usage within a typical workflow is described here.

jansodoge/awesom_dev_version documentation built on Jan. 26, 2021, 8:53 a.m.