Communication Analysis - Research Overview

Research Topic: The analysis of human communication in digital environments presents a complex, interdisciplinary challenge. Traditional approaches, often grounded in isolated paradigms - such as metadata extraction, social network analysis or NLP-based processes - struggle to account for the multimodal, contextual, and interwined nature of communication data. This research website presents a body of peer-reviewed articles and demonstrators situated at the intersection of computer science, digital humanities, and visual analytics, with the aim of advancing a formalized, semi-automated understanding for communication analysis in the age of modern AI solutions. The contributions showcased here explore how visual analytics systems can facilitate the identification, interpretation, and contextualization of communication behavior across heterogeneous modalities. By tightly coupling machine learning models with human expertise through iterative visual interfaces, this research addresses key methodological and ethical challenges - including data heterogeneity, user agency, and the need for explainability in sensitive investigative settings. Building upon conceptual foundations, state-of-the-art surveys, and applied system demonstrators, the presented work introduces techniques for communication pattern recognition, temporal hypergraph modeling, metadata-driven behavior analysis, and multimodal intelligence frameworks. Together, these approaches contribute to a more integrated, explainable, and efficient mode of digital communication analysis.

This research was conducted a the University of Konstanz, Germany, between 2019 and 2023, and forms part of the following dissertation: