Hapax and square-root corrections turn uncalibrated LambdaG scores into well-calibrated likelihood ratios for forensic authorship verification, matching logistic regression on fifteen corpora without training a calibration model.
Authorship Verification as a One-Class Classification Problem , booktitle =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Normalisation-Based Likelihood Ratio Estimation for Forensic Authorship Verification
Hapax and square-root corrections turn uncalibrated LambdaG scores into well-calibrated likelihood ratios for forensic authorship verification, matching logistic regression on fifteen corpora without training a calibration model.