NETANOS - Named entity-based Text Anonymization for Open Science

Background: The shift towards open science, implies that researchers should share their data. Often there is a dilemma between publicly sharing data and protecting their subjects' confidentiality. Moreover, the case of unstructured text data (e. g. stories) poses an additional dilemma: anonymizing texts without deteriorating their content for secondary research. Existing text anonymization systems either deteriorate the content of the original or have not been tested empirically. We propose and empirically evaluate NETANOS: named entity-based text anonymization for open science. NETANOS is an open-source context-preserving anonymization system that identifies and modifies named entities (e. g. persons, locations, times, dates). The aim is to assist researchers in sharing their raw text data. Method Results: NETANOS anonymizes critical, contextual information through a stepwise named entity recognition (NER) implementation: it identifies contextual information (e. g. "Munich") and then replaces them with a context-preserving category label (e. g. "Location₁"). We assessed how good participants were in re-identifying several travel stories (e. g. locations, names) that were presented in the original (“Max”), human anonymized (“Max” → “Person1”), NETANOS (”Max” → “Person1”), and in a context-deteriorating state (“Max” → “XXX”). Bayesian testing revealed that the NETANOS anonymization was practically equivalent to the human baseline anonymization. Conclusions: Named entity recognition can be applied to the anonymization of critical, identifiable information in text data. The proposed stepwise anonymization procedure provides a fully automated, fast system for text anonymization. NETANOS might be an important step to address researchers' dilemmas when sharing text data within the open science movement.