Transition Network Analysis in the browser
This is an interactive demo of tnaj, a pure TypeScript/JavaScript implementation of Transition Network Analysis (TNA). TNA is a novel framework for modeling, visualizing, and identifying temporal patterns in sequential data. It represents sequences as weighted directed networks where nodes are states and edges are transitions, enabling the application of network science methods (centralities, community detection, pruning) to understand the structure and dynamics of sequential processes.
The package runs entirely in the browser with zero server-side dependencies. It is a port of the R TNA package and the Python tna package, validated for numerical equivalence with R to machine epsilon (~10-15). Use the controls below to explore different model types, pruning thresholds, and community detection methods.