Determining an author's native language by mining a text for errors
Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Foundations and Trends in Information Retrieval
Automatically profiling the author of an anonymous text
Communications of the ACM - Inspiring Women in Computing
Computational methods in authorship attribution
Journal of the American Society for Information Science and Technology
A survey of modern authorship attribution methods
Journal of the American Society for Information Science and Technology
Empirical evaluation of authorship obfuscation using JGAAP
Proceedings of the 3rd ACM workshop on Artificial intelligence and security
Detecting Hoaxes, Frauds, and Deception in Writing Style Online
SP '12 Proceedings of the 2012 IEEE Symposium on Security and Privacy
| Hi-index | 0.00 |
Whistleblowers and activists need the ability to communicate without disclosing their identity, as of course do kidnappers and terrorists. Recent advances in the technology of stylometry (the study of authorial style) or "authorship attribution" have made it possible to identify the author with high reliability in a non-confrontational setting. In a confrontational setting, where the author is deliberately masking their identity (i.e. attempting to deceive), the results are much less promising. In this paper, we show that although the specific author may not be identifiable, the intent to deceive and to hide his identity can be. We show this by a reanalysis of the Brennan and Greenstadt (2009) deception corpus and discuss some of the implications of this surprising finding.