stCEG: An R Package for Modelling Events over Spatial Areas Using Chain Event Graphs

2025-07-09

AI Summary

Chain Event Graphs (CEGs) lack accessible, scalable software implementations, hindering their adoption for real-world event process modeling. Method: We developed stCEG, the first R package supporting end-to-end customizable CEG modeling—including data-driven structure learning, parameter estimation, and interactive visualization—grounded in Chain Event Graph theory. Built atop Shiny, it provides a no-code web-based GUI. Contribution/Results: stCEG introduces the first explicit integration of spatial region variables into the CEG framework, enabling representation and exploration of spatial heterogeneity in event pathways. Applied to London homicide data, it significantly enhances CEGs’ interpretability, interactivity, and practical utility for complex social event analysis, thereby extending the applicability of CEGs to spatial event modeling.

Technology Category

• Reasoning under Uncertainty: Graphical Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data visualization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

stCEG is an R package which allows a user to fully specify a Chain Event Graph (CEG) model from data and to produce interactive plots. It includes functions for the user to visualise spatial variables they wish to include in the model. There is also a web-based graphical user interface (GUI) provided, increasing ease of use for those without knowledge of R. We demonstrate stCEG using a dataset of homicides in London, which is included in the package. stCEG is the first software package for CEGs that allows for full model customisation.

Problem

• Develops R package for spatial event modeling using Chain Event Graphs
• Enables interactive visualization of spatial variables in CEG models
• Provides GUI for non-R users to customize CEG models easily

Innovation

• R package for spatial Chain Event Graphs
• Interactive plots and web-based GUI
• Full model customization for CEGs