Logo image
Back
A Girl Has A Name: Detecting Authorship Obfuscation
Conference proceeding   Open access

A Girl Has A Name: Detecting Authorship Obfuscation

Asad Mahmood, Zubair Shafiq and Padmini Srinivasan
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp.2235-2245
07/2020
DOI: 10.18653/v1/2020.acl-main.203
url
https://doi.org/10.18653/v1/2020.acl-main.203View
Published (Version of record) Open Access

Abstract

Authorship attribution aims to identify the author of a text based on the stylometric analysis. Authorship obfuscation, on the other hand, aims to protect against authorship attribution by modifying a text’s style. In this paper, we evaluate the stealthiness of state-of-the-art authorship obfuscation methods under an adversarial threat model. An obfuscator is stealthy to the extent an adversary finds it challenging to detect whether or not a text modified by the obfuscator is obfuscated – a decision that is key to the adversary interested in authorship attribution. We show that the existing authorship obfuscation methods are not stealthy as their obfuscated texts can be identified with an average F1 score of 0.87. The reason for the lack of stealthiness is that these obfuscators degrade text smoothness, as ascertained by neural language models, in a detectable manner. Our results highlight the need to develop stealthy authorship obfuscation methods that can better protect the identity of an author seeking anonymity.

Details

Metrics

42 Record Views
Logo image