Features and supported methods

Generate unsupervised labels by identifying structure in feature space or an embedded representation.

Train and evaluate models using a known target for classification or regression.

Construct lower-dimensional representations for visualization, comparison, and embedded-space clustering.

Assess runs using standardized metrics and higher-level analytical pillars.

Test whether an analytical result persists when the underlying data or analytical conditions are deliberately perturbed.

Use SHAP-based feature importance to inspect which variables most strongly contribute to supervised or labeling results.

Generate compact rules that describe how a discovered label differs from other observations.

Investigate whether an existing analytical group contains meaningful internal structure.

Examine how labels, states, and analytical trajectories change across repeated observations.