This paper firstly proposes a simple yet efficient generalized approach to apply differential privacy to text representation (i.e., word embedding). Based on it, we propose a user-level approach to learn personalized differentially private word embedding model on user generated contents (UGC). To our best knowledge, this is the first work of learning user-level differentially private word embedding model from text for sharing. The proposed approaches protect the privacy of the individual from re-identification, especially provide better trade-off of privacy and data utility on UGC data for sharing. The experimental results show that the trained embedding models are applicable for the classic text analysis tasks (e.g., regression). Moreover, the proposed approaches of learning difierentially private embedding models are both framework- and dataindependent, which facilitates the deployment and sharing. The source code is available at https://github.com/sonvx/dpText.

Item Type:	Refereed Conference Paper
Keywords:	privacy, generated text, private word embedding, differential privacy, UGC
Research Division:	Information and Computing Sciences
Research Group:	Artificial intelligence
Research Field:	Natural language processing
Objective Division:	Information and Communication Services
Objective Group:	Information systems, technologies and services
Objective Field:	Information systems, technologies and services not elsewhere classified
UTAS Author:	Tran, SN (Dr Son Tran)
ID Code:	139115
Year Published:	2019
Deposited By:	Information and Communication Technology
Deposited On:	2020-05-27
Last Modified:	2020-06-16
Downloads:	0