Shifting Dataset to Preserve Data Privacy
Data analytic is very valuable in any domain that produces large amount of data making demands on full datasets to be revealed for analytic purposes are rising. Regardless, the privacy of the released dataset should be preserved. New techniques using synthetic data as a mean to preserve the privacy...
保存先:
| 主要な著者: | , , , , , |
|---|---|
| フォーマット: | Conference Paper |
| 言語: | English |
| 出版事項: |
2020
|
| タグ: |
タグ追加
タグなし, このレコードへの初めてのタグを付けませんか!
|
| 要約: | Data analytic is very valuable in any domain that produces large amount of data making demands on full datasets to be revealed for analytic purposes are rising. Regardless, the privacy of the released dataset should be preserved. New techniques using synthetic data as a mean to preserve the privacy has been identified as appropriate approach to fulfill the demand. In this paper, a privacy-preserving data synthetic framework for data analytic is proposed. Using a generative model that captures the density function of data attributes, the privacy-preserving synthetic data is produced. We performed classification task through various machine learning classifiers in measuring the data utility of the new privacy-preserving synthesized data. © 2018 IEEE. |
|---|