Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

evomap: A Toolbox for Dynamic Mapping in Python

Maximilian Matthe

TL;DR

evomap delivers a comprehensive Python toolkit for dynamic mapping based on the EvoMap framework, bridging static mapping methods (MDS, Sammon, t-SNE) with temporal regularization to produce coherent trajectories over time. The package emphasizes modular design, scikit-learn–style APIs, end-to-end workflows (preprocessing, fitting, visualization, evaluation), and hyperparameter optimization (grid search and Bayesian methods) for balancing static fit with temporal alignment. Through a detailed TNIC usage example and simulated-data experiments, the authors demonstrate improved temporal coherence, robust recovery of latent trajectories, and scalable performance relative to static approaches. These contributions enable researchers to visualize and quantify evolving relational structures across diverse domains, with practical guidance for handling unbalanced data and tuning hyperparameters. The work lays a foundation for extending EvoMap to additional mapping methods and broader types of dynamic relational data.

Abstract

This paper presents evomap, a Python package for dynamic mapping. Mapping methods are widely used across disciplines to visualize relationships among objects as spatial representations, or maps. However, most existing statistical software supports only static mapping, which captures objects' relationships at a single point in time and lacks tools to analyze how these relationships evolve. evomap fills this gap by implementing the dynamic mapping framework EvoMap, originally proposed by Matthe, Ringel, and Skiera (2023), which adapts traditional static mapping methods for dynamic analyses. The package supports multiple mapping techniques, including variants of Multidimensional Scaling (MDS), Sammon Mapping, and t-distributed Stochastic Neighbor Embedding (t-SNE). It also includes utilities for data preprocessing, exploration, and result evaluation, offering a comprehensive toolkit for dynamic mapping applications. This paper outlines the foundations of static and dynamic mapping, describes the architecture and functionality of evomap, and illustrates its application through an extensive usage example.

evomap: A Toolbox for Dynamic Mapping in Python

TL;DR

evomap delivers a comprehensive Python toolkit for dynamic mapping based on the EvoMap framework, bridging static mapping methods (MDS, Sammon, t-SNE) with temporal regularization to produce coherent trajectories over time. The package emphasizes modular design, scikit-learn–style APIs, end-to-end workflows (preprocessing, fitting, visualization, evaluation), and hyperparameter optimization (grid search and Bayesian methods) for balancing static fit with temporal alignment. Through a detailed TNIC usage example and simulated-data experiments, the authors demonstrate improved temporal coherence, robust recovery of latent trajectories, and scalable performance relative to static approaches. These contributions enable researchers to visualize and quantify evolving relational structures across diverse domains, with practical guidance for handling unbalanced data and tuning hyperparameters. The work lays a foundation for extending EvoMap to additional mapping methods and broader types of dynamic relational data.

Abstract

This paper presents evomap, a Python package for dynamic mapping. Mapping methods are widely used across disciplines to visualize relationships among objects as spatial representations, or maps. However, most existing statistical software supports only static mapping, which captures objects' relationships at a single point in time and lacks tools to analyze how these relationships evolve. evomap fills this gap by implementing the dynamic mapping framework EvoMap, originally proposed by Matthe, Ringel, and Skiera (2023), which adapts traditional static mapping methods for dynamic analyses. The package supports multiple mapping techniques, including variants of Multidimensional Scaling (MDS), Sammon Mapping, and t-distributed Stochastic Neighbor Embedding (t-SNE). It also includes utilities for data preprocessing, exploration, and result evaluation, offering a comprehensive toolkit for dynamic mapping applications. This paper outlines the foundations of static and dynamic mapping, describes the architecture and functionality of evomap, and illustrates its application through an extensive usage example.

Paper Structure

This paper contains 27 sections, 6 equations, 12 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Background on mapping
  3. Static mapping
  4. Dynamic mapping
  5. The evomap package
  6. Installation
  7. Key dependencies
  8. Package overview
  9. Package design and implementation
  10. Detailed usage example
  11. Preprocessing
  12. Fitting EvoMap to data
  13. Detailed setup of EvoMap
  14. Starting positions
  15. EvoMap's hyperparameters
  16. ...and 12 more sections

Figures (12)

  • Figure 1: Illustration of dynamic mapping via EvoMap
  • Figure 2: Overview of the mapping module
  • Figure 3: Dynamic map of nine technology firms, 1998-2017
  • Figure 4: Dynamic map under different hyperparameter choices
  • Figure 5: Three static EvoMap snapshots
  • ...and 7 more figures