ABOUT

ER Visual Lab is a browser-based engine for scientific network visualization, built from first principles in TypeScript with no runtime dependencies. It transforms relational data—including edge lists, adjacency matrices, GraphML, Pajek, and two-mode or multilevel tables—into interactive, publication-quality visualizations, rendered using either an HTML5 canvas or a GPU-accelerated WebGL pipeline capable of scaling to networks with hundreds of thousands of nodes and edges.

More than a visualization tool, ER Visual Lab computes the structure it displays. It includes state-of-the-art algorithms for community detection (Louvain, Leiden, Infomap, Label Propagation, and Girvan–Newman), centrality analysis (degree, betweenness, closeness, eigenvector, PageRank, and k-core), and a comprehensive collection of layouts, ranging from force-directed methods (Fruchterman–Reingold, Kamada–Kawai, and stress majorization) to concentric ego and centrality views, geographic projections, and true multilevel visualizations with fully rotatable 3D planes.

Every visual attribute—node size, shape, colour, borders, edge width, opacity, and style—can be mapped directly to data, while figures can be exported as SVG, PNG, or self-contained interactive HTML for seamless embedding in websites, presentations, and publications. Designed for teaching, research, and scientific communication in network science, ER Visual Lab aims not to replicate existing software, but to provide a rigorous, elegant, and fully controllable visual language for exploring complex relational systems.

ER Visual Lab will be released soon.