Overview

This documentation provides an overview on how to use self-organizing map with relational perspective mapping (SOM-RPM) toolbox for MATLAB.

SOM-RPM is an unsupervised machine learning algorithm designed to model and visualise both the topological and distance information in a data set. The SOM part of the algorithm is responsible for modelling topology, while RPM handles the modelling of distances.

SOM-RPM was developed for hyperspectral data, specifically time-of-flight secondary ion mass spectrometry (ToF-SIMS) images and depth profiles. This toolbox is built around this use case, with specific functionality designed for interactive exploration of hyperspectral data. The remainder of these documentations explains how to best utilize the SOM-RPM toolbox to this end.

Our primary aim is to make the SOM-RPM method more accessible to the broader scientific community. This documentation provides sufficient content for a non-expert in machine learning to be able to utilize SOM-RPM for exploratory analysis of their hyperspectral data.

For more information about SOM-RPM, please refer to the original SOM-RPM paper, or to the paper associated with the release of this toolbox.

Documentation Layout

The panel on the left-hand side of the screen contains everything you'll need to get started with the SOM-RPM toolbox and to use as a reference while you work. This includes information on getting started/installation, the toolbox structure, general usage guidelines and a tutorial-style case study. You can follow along with the analysis in this case study using the embedded coded examples and the example data set included with the toolbox.

If you use this toolbox or our data in your work, we ask that you please cite our toolbox. For information on how to do this, refer to the attribution and licence tabs.