Chapter III Judicial Power Corpus — Pilot Dashboard
Testing whether dictionary-based corpus analysis can track the expansion, restraint, and doctrinal development of judicial power in Australian constitutional cases.
Corpus Metadata
The current pilot scans published judgment text only. Oral argument transcripts and speaker turns should be treated as a separate corpus if added later.
Dictionary Signals
Rates are normalised per 100,000 words across the 20-case pilot. They indicate dictionary behaviour in this test set, not a historical trend.
Dictionary Performance Review
These tables turn KWIC review into a diagnostic instrument: which dictionaries are plausible, which terms are noisy, and where review effort should go next.
Case Comparison
The expansion score is a composite diagnostic from the v0 dictionaries. It is a prioritisation aid for review, not a finding about doctrinal change.
Dictionary-By-Case Heatmap
Cells are column-scaled, so each dictionary can show whether expected cases light up against its own distribution.
KWIC Context Review
Manual validation is the academic control layer: accepted, rejected, and uncertain judgments should be reviewed before scaling.
Pilot Charts
These charts help evaluate dictionary behaviour and case-level signal. The year chart is labelled as pilot distribution only.
Pilot Findings
data/processed/chapter_iii_v0. On the live Cloudflare dashboard, KWIC validation dropdowns are shared through the review API; local file views fall back to this browser only.