Factor Network

About this visualization
This visualization represents a 3-dimensional projection of a Factor Analysis (FA) of melodic features for the Essen Folksong Collection (Schaffrath, 1995).

Factor Analysis is a statistical method that seeks to identify underlying latent factors that explain the correlations between variables. This allows us to reduce the dimensionality of the data while preserving the underlying structure, and to relate this underlying structure to psychological or cognitive constructs through our interpretation of the factors.

In this example, we have identified 17 factors that explain the correlations between the 189 numerical features utilised in this FA solution. By visualising the network of factor loadings, we can see which features are most strongly associated with each factor. This facilitates our interpretation of these factors. For instance, we can clearly see that Factor 1 is strongly associated with mostly Timing features, so we assign this factor the name "Timing". We can apply this approach to all 17 factors to assign names to each factor.

The 3D network diagram provides an exciting interactive environment for exploring the different relationships between the factors and features. We can isolate feature nodes, factors, and move them around to traverse the complex interrelationships between them.

This visualization runs using the Three.js library for 3D rendering, and the 3D Force-Directed Graph library for the interactive network diagram. Features were calculated using melody-features, a Python package for analysing monophonic MIDI melodies (Whyatt and Harrison, 2025). These features were computed on all melodies in the Essen Folksong Collection. melody-features can be found at and cited using the following DOI: doi.org/10.5281/zenodo.16894207