Abstract

Most of real data is imbalanced data and it is difficult data as a subject of classification problem. For imbalanced data, there are some researches. However, handling of imbalance data in deep learning has not been studied. In this paper, we propose privacy protection data mining using imbalance data through deep learning. We discuss existing privacy protection data mining, study its features, and examine an anonymizing tool for deep learning. Experiments using anonymization tools (UAT) confirmed that deep learning does not reduce accuracy by
making it anonymous. Undersampling and upper sampling are used for imbalance data, but it was not confirm the tendency of lowering accuracy due to anonymization.

Published in: Internet Technology and Secured Transactions (ICITST-2018)

  • Date of Conference: 10-13 December 2018
  • DOI: 10.2053/ICITST.WorldCIS.WCST.WCICSS.2018.0014
  • ISBN: 978-1-908320-94-0
  • Conference Location: University of Cambridge, Churchill College