Towards a private vector space model for confidential documents

We introduce in this paper a method to anonymize document vector spaces. These vector spaces can be used to analyze confidential documents without disclosing private information. The method is inspired in microaggregation, a popular technique used in statistical disclosure control. URL http://doi.acm.org/10.1145/2480362.2480543 [9] Source URL: https://www.iiia.csic.es/en/node/54488 Links [1] https://www.iiia.csic.es/en/staff/daniel-abril [2] https://www.iiia.csic.es/en/staff/guillermo-navarro-arribas [3] https://www.iiia.csic.es/en/staff/vicen%C3%A7-torra [4] https://www.iiia.csic.es/en/bibliography?f[author]=699 [5] https://www.iiia.csic.es/en/bibliography?f[keyword]=565 [6] https://www.iiia.csic.es/en/bibliography?f[keyword]=566 [7] https://www.iiia.csic.es/en/bibliography?f[keyword]=567 [8] https://www.iiia.csic.es/en/bibliography?f[keyword]=461 [9] http://doi.acm.org/10.1145/2480362.2480543