TY - GEN
T1 - A Word is Worth A Thousand Dollars
T2 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022
AU - Xie, Yong
AU - Wang, Dakuo
AU - Chen, Pin Yu
AU - Xiong, Jinjun
AU - Liu, Sijia
AU - Koyejo, Sanmi
N1 - Publisher Copyright:
© 2022 Association for Computational Linguistics.
PY - 2022
Y1 - 2022
N2 - More and more investors and machine learning models rely on social media (e.g., Twitter and Reddit) to gather real-time information and sentiment to predict stock price movements. Although text-based models are known to be vulnerable to adversarial attacks, whether stock prediction models have similar vulnerability is underexplored. In this paper, we experiment with a variety of adversarial attack configurations to fool three stock prediction victim models. We address the task of adversarial generation by solving combinatorial optimization problems with semantics and budget constraints. Our results show that the proposed attack method can achieve consistent success rates and cause significant monetary loss in trading simulation by simply concatenating a perturbed but semantically similar tweet.
AB - More and more investors and machine learning models rely on social media (e.g., Twitter and Reddit) to gather real-time information and sentiment to predict stock price movements. Although text-based models are known to be vulnerable to adversarial attacks, whether stock prediction models have similar vulnerability is underexplored. In this paper, we experiment with a variety of adversarial attack configurations to fool three stock prediction victim models. We address the task of adversarial generation by solving combinatorial optimization problems with semantics and budget constraints. Our results show that the proposed attack method can achieve consistent success rates and cause significant monetary loss in trading simulation by simply concatenating a perturbed but semantically similar tweet.
UR - http://www.scopus.com/inward/record.url?scp=85138432590&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85138432590&partnerID=8YFLogxK
U2 - 10.18653/v1/2022.naacl-main.43
DO - 10.18653/v1/2022.naacl-main.43
M3 - Conference contribution
AN - SCOPUS:85138432590
T3 - NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference
SP - 587
EP - 599
BT - NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics
PB - Association for Computational Linguistics (ACL)
Y2 - 10 July 2022 through 15 July 2022
ER -