The Potential for Misleading Results in Text Sanitisation with Standard Evaluation Metrics

Dan Zhang, Mark Anderson


Abstract
Data privacy is an important facet of modern life. It is especially important when considering data that carries potentially sensitive information such as in medical or legal documents. However, it is particularly difficult to ensure private information has been removed or masked in unstructured data, e.g. free-flowing text. The evaluation of systems that automatically detect and remove personal identifiable information (PII) from text is also challenging. Here we present a case study of a system that seemingly performed well, but under closer scrutiny the high performance was due to the shortcomings of standard binary classification metrics in the context of high target class prevalence. We then give a short analysis of different possible metrics in these high-prevalence scenarios, clearly showing the superiority of the Matthews Correlation Coefficient. This is particularly important because readily available data in this domain is rare and often systems are compared using biographies from Wikipedia which have a naturally high prevalence. This can be further aggravated by certain reasonable pre-processing or evaluation formalisms as in the case study discussed here.
Anthology ID:
2026.lrec-1.364
Volume:
Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:
LREC
SIG:
Publisher:
ELRA Language Resource Association
Note:
Pages:
4638–4646
Language:
External URL:
https://lrec.elra.info/lrec2026-main-364
DOI:
10.63317/4ubbuzpc4hpu
Bibkey:
Cite (ACL):
Dan Zhang and Mark Anderson. 2026. The Potential for Misleading Results in Text Sanitisation with Standard Evaluation Metrics. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 4638–4646, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):
The Potential for Misleading Results in Text Sanitisation with Standard Evaluation Metrics (Zhang & Anderson, LREC 2026)
Copy Citation: