Uzbek MT structure analysis: clause-model hierarchy and measurement-validity diagnostics (code and coded data)

Reproducibility package for two studies on Uzbek sentence structure and rule-based MT diagnostics. hierarchy/ — clause-model inventory and Heaps' law (β) measurement across three levels of analysis (word, block, clause), control experiments, cross-corpus coverage. diagnostics/ — pipeline tracer (mirror method), failure taxonomy, external gold standard (2 annotators + arbitration), coding validation (Cohen's κ = 0.822). Contains code and coded data only; raw corpus texts are not included (see README for the rationale). Code: MIT. Data and docs: CC BY 4.0.

Hali tarjima qilinmagan