Towards a Human-AI Hybrid for Adversarial Authorship

In this paper, we compare two types of masking methods for Adversarial Authorship. One method is a human-based interactive form of masking (referred to as AuthorCAAT-V) while the second method is hybrid of three state-of-the-art author masking techniques (referred to as AIM-IT). Our results show that the performances of AuthorCAAT-V and AIM-IT are equal to or better than the performances of the three state-of-the-art author masking techniques in reducing the identification rate of four well-known authorship attribution systems (AASs). Furthermore, our results show that the hybridization of AuthorCAAT-V and AIM-IT provides a greater reduction in the identification rate.

[1]  Jinsheng Xu,et al.  Adversarial authorship, interactive evolutionary hill-climbing, and author CAAT-III , 2017, 2017 IEEE Symposium Series on Computational Intelligence (SSCI).

[2]  Daniel Castro-Castro,et al.  Author Masking by Sentence Transformation , 2017, CLEF.

[3]  Shlomo Argamon,et al.  Authorship attribution in the wild , 2010, Lang. Resour. Evaluation.

[4]  George A. Miller,et al.  WordNet: A Lexical Database for English , 1995, HLT.

[5]  Ismail Kassou,et al.  Authorship Analysis Studies: A Survey , 2014 .

[6]  Matthias Hagen,et al.  Overview of the Author Obfuscation Task at PAN 2018: A New Approach to Measuring Safety , 2018, CLEF.

[7]  Lawrence. Davis,et al.  Handbook Of Genetic Algorithms , 1990 .

[8]  Damon L. Woodard,et al.  GEFeS: Genetic & evolutionary feature selection for periocular biometric recognition , 2011, 2011 IEEE Workshop on Computational Intelligence in Biometrics and Identity Management (CIBIM).

[9]  William John Teahan,et al.  A repetition based measure for verification of text collections and for text categorization , 2003, SIGIR.

[10]  Rachel Greenstadt,et al.  Anonymouth Revamped : Getting Closer to Stylometric Anonymity , 2013 .

[11]  Mohammad Hossein Jarrahi,et al.  Artificial intelligence and the future of work: Human-AI symbiosis in organizational decision making , 2018, Business Horizons.

[12]  Michael Gamon,et al.  Obfuscating Document Stylometry to Preserve Author Anonymity , 2006, ACL.

[13]  Oleg Bakhteev,et al.  Author Masking using Sequence-to-Sequence Models , 2017, CLEF.

[14]  Kaushik Roy,et al.  Genetic & Evolutionary Feature Selection for Author Identification of HTML Associated with Malware , 2014 .

[15]  J. Pennebaker,et al.  The Psychological Meaning of Words: LIWC and Computerized Text Analysis Methods , 2010 .

[16]  Mostafa Rahgouy,et al.  Author Masking Directed by Author's Style: Notebook for PAN at CLEF 2018 , 2018, CLEF.

[17]  Ellen Riloff,et al.  Learning subjective nouns using extraction pattern bootstrapping , 2003, CoNLL.

[18]  Rachel Greenstadt,et al.  Practical Attacks Against Authorship Recognition Techniques , 2009, IAAI.

[19]  Jinsheng Xu,et al.  Authorship Attribution vs. Adversarial Authorship from a LIWC and Sentiment Analysis Perspective , 2018, 2018 IEEE Symposium Series on Computational Intelligence (SSCI).

[20]  Jinsheng Xu,et al.  Authorship Attribution via Evolutionary Hybridization of Sentiment Analysis, LIWC, and Topic Modeling Features , 2018, 2018 IEEE Symposium Series on Computational Intelligence (SSCI).

[21]  Gerry Dozier,et al.  The Best Way to a Strong Defense is a Strong Offense : Mitigating Deanonymization Attacks via Iterative Language Translation , .

[22]  James Brown,et al.  Adversarial Authorship, AuthorWebs, and Entropy-Based Evolutionary Clustering , 2016, 2016 25th International Conference on Computer Communication and Networks (ICCCN).

[23]  Fuchun Peng,et al.  N-GRAM-BASED AUTHOR PROFILES FOR AUTHORSHIP ATTRIBUTION , 2003 .

[24]  Ariel Stolerman,et al.  Use Fewer Instances of the Letter "i": Toward Writing Style Anonymization , 2012, Privacy Enhancing Technologies.

[25]  Taher Rahgooy,et al.  obfuscation using WordNet and language models Notebook for PAN at CLEF 2016 , 2016 .

[26]  Claire Cardie,et al.  Annotating Expressions of Opinions and Emotions in Language , 2005, Lang. Resour. Evaluation.

[27]  Gerry V. Dozier,et al.  Adversarial Authorship, Sentiment Analysis, and the AuthorWeb Zoo , 2018, 2018 IEEE Symposium Series on Computational Intelligence (SSCI).

[28]  Ellen Riloff,et al.  Learning Extraction Patterns for Subjective Expressions , 2003, EMNLP.

[29]  Paolo Rosso,et al.  Convolutional Neural Networks for Authorship Attribution of Short Texts , 2017, EACL.

[30]  Yiming Yan,et al.  Surveying Stylometry Techniques and Applications , 2017, ACM Comput. Surv..

[31]  Janyce Wiebe,et al.  Recognizing Contextual Polarity in Phrase-Level Sentiment Analysis , 2005, HLT.

[32]  Jacques Savoy,et al.  UniNE at CLEF 2018: Author Masking: Notebook for PAN at CLEF 2018 , 2018, CLEF.