Node Reduction through Artificial Intelligence Inferences using Graphology and SigmaJS: A Case Study on Hypertextual Conversations in Freight Train Graffiti in the North American Region

Lecture presented at the international conference FOSDEM 2024 (Brussels, Belgium). Presents a technical front-end web development solution to solve the 'hairball problem' (visual clutter and node overcrowding) in dense relational graphs.

Using Instagram hashtag networks as a case study, the presentation demonstrates how to implement interactive spatial clustering algorithms in the browser. The logic, written in JavaScript with Graphology and rendered on WebGL with SigmaJS, dynamically reduces peripheral nodes based on structural centrality weights, facilitating real-time visual analysis on standard consumer devices.