2020/03/06 by Aitor García Pablos, García-Pablos, Aitor, Naiara Pérez +3
Computer Science · Decision Sciences · #Computation and Language (cs.CL) #Data Quality and Management #FOS: Computer and information sciences #Hate Speech and Cyberbullying Detection #Natural Language Processing Techniques #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2003.03106
openalex publication_date 2020/03/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Massive digital data processing provides a wide range of opportunities and\nbenefits, but at the cost of endangering personal data privacy. Anonymisation\nconsists in removing or replacing sensitive information from data, enabling its\nexploitation for different purposes while preserving the privacy of\nindividuals. Over the years, a lot of automatic anonymisation systems have been\nproposed; however, depending on the type of data, the target language or the\navailability of training documents, the task remains challenging still. The\nemergence of novel deep-learning models during the last two years has brought\nlarge improvements to the state of the art in the field of Natural Language\nProcessing. These advancements have been most noticeably led by BERT, a model\nproposed by Google in 2018, and the shared language models pre-trained on\nmillions of documents. In this paper, we use a BERT-based sequence labelling\nmodel to conduct a series of anonymisation experiments on several clinical\ndatasets in Spanish. We also compare BERT to other algorithms. The experiments\nshow that a simple BERT-based model with general-domain pre-training obtains\nhighly competitive results without any domain specific feature engineering.\n