rolescry-package: rolescry: Name-blind Variable-Role Detection

In rolescry: Name-Blind Variable-Role Detection by Data Signature

rolescry: Name-blind Variable-Role Detection

Description

Deterministic, name-blind detection of variable roles (group, outcome, survival time/event, paired, agreement, repeated measures, scale items, subject id, covariate) in tabular data, using information-theoretic signatures (Shannon entropy, normalized mutual information) and distributional shape rather than column names. The guiding principle is "Data inspice, non nomen" – inspect the data, not the name.

Details

The single public entry point is detect_roles(). Header-aware data loading is available via read_data(). No LLMs, no external data transmission; detection is >= 90 percent mathematical signature with an optional, capped (<= 10 percent) name bonus (see the name_bonus argument).

Developed from work on the author's unreleased MDStatR research code; see the README section "Relationship to MDStatR". rolescry requires only base R and has no MDStatR dependency.

Author(s)

Maintainer: Can Boynukara [e-mail] (ORCID) [copyright holder]

See Also

Useful links:

• https://github.com/canboynukara/rolescry

• Report bugs at https://github.com/canboynukara/rolescry/issues

rolescry documentation built on Oct. 21, 2026, 5:14 p.m.