Search papers, labs, and topics across Lattice.
The authors introduce MultiGraSCCo, a multilingual anonymization benchmark comprising synthetic medical text in ten languages, designed for training and evaluating anonymization systems. They use machine translation, adapting personal identifiers to be culturally and contextually appropriate in each language, while preserving original annotations. Evaluation by medical professionals confirms the quality of the translated data and adapted personal information, making the benchmark suitable for various applications like training annotators and improving PII detection.
A new multilingual benchmark dataset with over 2,500 annotations of personal information enables privacy-preserving machine learning across ten languages, sidestepping the need for sensitive patient data.
Accessing sensitive patient data for machine learning is challenging due to privacy concerns. Datasets with annotations of personally identifiable information are crucial for developing and testing anonymization systems to enable safe data sharing that complies with privacy regulations. Since accessing real patient data is a bottleneck, synthetic data offers an efficient solution for data scarcity, bypassing privacy regulations that apply to real data. Moreover, neural machine translation can help to create high-quality data for low-resource languages by translating validated real or synthetic data from a high-resource language. In this work, we create a multilingual anonymization benchmark in ten languages, using a machine translation methodology that preserves the original annotations and renders names of cities and people in a culturally and contextually appropriate form in each target language. Our evaluation study with medical professionals confirms the quality of the translations, both in general and with respect to the translation and adaptation of personal information. Our benchmark with over 2,500 annotations of personal information can be used in many applications, including training annotators, validating annotations across institutions without legal complications, and helping improve the performance of automatic personal information detection. We make our benchmark and annotation guidelines available for further research.
