: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
arXiv:1711.01921
Abstract
Text-based analysis methods allow to reveal privacy relevant author attributes such as gender, age and identify of the text's author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation (), to combat such text-based adversaries. We combine sequence-to-sequence language models used in machine translation and generative adversarial networks to obfuscate author attributes. Unlike machine translation techniques which need paired data, our method can be trained on unpaired corpora of text containing different authors. Importantly, we propose and evaluate techniques to impose constraints on our to preserve the semantics of the input text. learns to make minimal changes to the input text to successfully fool author attribute classifiers, while aiming to maintain the meaning of the input. We show through experiments on two different datasets and three settings that our proposed method is effective in fooling the author attribute classifiers and thereby improving the anonymity of authors.
16 pages, 10 figures and 8 tables
References in corpus (7)
- Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- Style Transfer from Non-Parallel Text by Cross-Alignment
- Adversarial Learning for Neural Dialogue Generation
- Character-level and Multi-channel Convolutional Neural Networks for Large-scale Authorship Attribution
- Sequence-to-Sequence Models Can Directly Translate Foreign Speech
- Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training