v0.3.0

Added

• EDUCATIONAL content domain for educational platforms, MOOCs, open textbooks, and curriculum bodies
• Domain prior: EDUCATIONAL=0.65, threshold=0.50, bayesianThreshold=0.65
• URL patterns: Khan Academy, OpenStax, Coursera, edX, Brilliant, CK-12, NCTM, and more
• Classification priority: ACADEMIC > NEWS > GOVERNMENT > EDUCATIONAL > GENERAL

Install

npm install groundcheck

v0.2.0

Added

• computeBayesianScore(): v2 Bayesian log-odds scoring with per-layer sensitivity/specificity
• LayerBayesianParams type: { sensitivity, specificity } per layer
• BayesianLayerConfig type: extends LayerConfig with a bayesian field
• prior field on DomainConfig: P(reference is real given domain)
• bayesianThreshold field on DomainConfig: minimum posterior to VERIFY
• Domain priors: ACADEMIC=0.72, NEWS=0.75, GOVERNMENT=0.82, GENERAL=0.45
• Bayesian thresholds: ACADEMIC=0.82, NEWS=0.65, GOVERNMENT=0.72, GENERAL=0.68
• 17 new Bayesian tests (126 total)

Changed

• DomainConfig.layers type widened from LayerConfig[] to BayesianLayerConfig[]
• DOMAIN_CONFIGS updated with calibrated sensitivity/specificity for all layers

Notes

• The v1 API (computeDomainAwareScore, DomainConfig.threshold) is fully backward-compatible.

v0.1.0

Initial release of the domain-aware reference verification standard.

Added

• classifyReference(): classify a reference into ACADEMIC, NEWS, GOVERNMENT, or GENERAL domain
• computeDomainAwareScore(): compute a weighted verification score against a domain's threshold
• DOMAIN_CONFIGS: full domain configuration map with layers, weights, thresholds, and AI instructions
• TypeScript types: ContentDomain, LayerId, LayerResult, LayerConfig, DomainConfig
• Comprehensive test suite (classify, score, domains)

Domain scoring formulas

• ACADEMIC: doi(0.45) + title_search(0.30) + url(0.10) + ai(0.15) >= 0.70
• NEWS: url(0.35) + ai(0.65) >= 0.50
• GOVERNMENT: url(0.40) + ai(0.60) >= 0.55
• GENERAL: url(0.30) + title_search(0.10) + ai(0.60) >= 0.55