$A^{4}NT$: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation

TL;DR

This paper introduces $A^4NT$, a neural translation-based adversarial method that obfuscates author attributes in text to enhance privacy, using unpaired data and semantic preservation techniques.

Contribution

The paper presents a novel adversarial training approach combining neural translation and GANs for author attribute anonymization without requiring paired datasets.

Findings

$A^4NT$ effectively fools author attribute classifiers.

It preserves input semantics while anonymizing authorship.

The method improves author anonymity across multiple datasets.

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 ($A^4NT$), 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 $A^4NT$ to preserve the semantics of the input text. $A^4NT$ 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.