Benjamin Zi Hao Zhao

Benjamin Zi Hao Zhao

Auto-generated profile, based on publicly available information

Benjamin Zi Hao Zhao

Macquarie University

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 September 2023 in Journal of Computer Virology and Hacking Techniques
Verified
Journal Article
Closed access
PDF file
HAHassan Jameel Asghar
Hassan Jameel Asghar
BZBenjamin Zi Hao Zhao
Benjamin Zi Hao Zhao
MIMuhammad Ikram
Muhammad Ikram
GNGiang Nguyen
Giang Nguyen
DKDali Kaafar
Dali Kaafar
DCDaniel Coscia
Daniel Coscia

27 February 2023 in IEEE Pervasive Computing
Verified
Journal Article
Open access
Benjamin TagBenjamin Tag
Benjamin Tag
NBNiels van Berkel
Niels van Berkel
SVSunny Verma
Sunny Verma
BZBenjamin Zi Hao Zhao
Benjamin Zi Hao Zhao
Berkovsky ShlomoBerkovsky Shlomo
Berkovsky Shlomo
OOOlga Ohrimenko
Olga Ohrimenko
Artificial intelligence (AI) systems have been increasingly used to make decision-making processes faster, more accurate, and more efficient. However, such systems are also at constant risk of being attacked. While the majority of attacks targeting AI-based applications aim to manipulate classifiers or training data and alter the output of an AI model, recently proposed sponge attacks against AI models aim to impede the classifier’s execution by consuming substantial resources. In this work, we propose dual denial of decision (DDoD) attacks against collaborative human-AI teams. We discuss how such attacks aim to deplete both computational and human resources, and significantly impair decision-making capabilities. We describe DDoD on human and computational resources and present potential risk scenarios in a series of exemplary domains.

30 April 2023
Verified
Conference Paper
Open access
HQHina Qayyum
Hina Qayyum
BZBenjamin Zi Hao Zhao
Benjamin Zi Hao Zhao
IWIan Wood
Ian Wood
MIMuhammad Ikram
Muhammad Ikram
NKNicolas Kourtellis
Nicolas Kourtellis
MKMohamad Ali Kaafar
Mohamad Ali Kaafar
Toxicity is endemic to online social networks (OSNs) including Twitter. It follows a Pareto-like distribution where most of the toxicity is generated by a very small number of profiles and as such, analyzing and characterizing these “toxic profiles” is critical. Prior research has largely focused on sporadic, event-centric toxic content (i.e., tweets) to characterize toxicity on the platform. Instead, we approach the problem of characterizing toxic content from a profile-centric point of view. We study 143K Twitter profiles and focus on the behavior of the top 1% producers of toxic content on Twitter, based on toxicity scores of their tweets availed by Perspective API. With a total of 293M tweets, spanning 16 years of activity, the longitudinal data allows us to reconstruct the timelines of all profiles involved. We use these timelines to gauge the behavior of the most toxic Twitter profiles compared to the rest of the Twitter population. We study the pattern of tweet posting from highly toxic accounts, based on the frequency and how prolific they are, the nature of hashtags and URLs, profile metadata, and Botometer scores. We find that the highly toxic profiles post coherent and well-articulated content, their tweets keep to a narrow theme with lower diversity in hashtags, URLs, and domains, they are thematically similar to each other, and have a high likelihood of bot-like behavior, likely to have progenitors with intentions to influence, based on high fake followers score. Our work contributes insight into the top 1% toxic profiles on Twitter and establishes the profile-centric approach to investigate toxicity on Twitter to be beneficial. The identification of the most toxic profiles can aid in the reporting and suspension of such profiles, making Twitter a better place for discussions. Finally, we contribute to the research community with this large-scale and longitudinal dataset1, annotated with six types of toxic scores.

1 April 2023 in Proceedings on Privacy Enhancing Technologies
Verified
Journal Article
Open access
PDF file
RARana Salal Ali
Rana Salal Ali
BZBenjamin Zi Hao Zhao
Benjamin Zi Hao Zhao
HAHassan Jameel Asghar
Hassan Jameel Asghar
TNTham Nguyen
Tham Nguyen
IWIan David Wood
Ian David Wood
MKMohamed Ali Kaafar
Mohamed Ali Kaafar
Named entity recognition models (NER), are widely used for identifying named entities (e.g., individuals, locations, and other information) in text documents. Machine learning based NER models are increasingly being applied in privacy-sensitive applications that need automatic and scalable identification of sensitive information to redact text for data sharing. In this paper, we study the setting when NER models are available as a black-box service for identifying sensitive information in user documents and show that these models are vulnerable to membership inference on their training datasets. With updated pre-trained NER models from spaCy [27], we demonstrate two distinct membership attacks on these models. Our first attack capitalizes on unintended memorization in the NER's underlying neural network, a phenomenon NNs are known to be vulnerable to. Our second attack leverages a timing side-channel to target NER models that maintain vocabularies constructed from the training data. We show that different functional paths of words within the training dataset in contrast to words not previously seen have measurable differences in execution time. Revealing membership status of training samples has clear privacy implications. For example, in text redaction, sensitive words or phrases to be found and removed, are at risk of being detected in the training dataset. Our experimental evaluation includes the redaction of both password and health data, presenting both security risks and a privacy/regulatory issues. This is exacerbated by results that indicate memorization after only a single phrase. We achieved a 70% AUC in our first attack on a text redaction use-case. We also show overwhelming success in the second timing attack with an 99.23% AUC. Finally we discuss potential mitigation approaches to realize the safe use of NER models in light of the presented privacy and security implications of membership inference attacks.

1 April 2025
Verified
Book Chapter
Closed access
PDF file
MSMuhammad Salman
Muhammad Salman
BZBenjamin Zi Hao Zhao
Benjamin Zi Hao Zhao
HAHassan Jameel Asghar
Hassan Jameel Asghar
MIMuhammad Ikram
Muhammad Ikram
SKSidharth Kaushik
Sidharth Kaushik
MKMohamed Ali Kaafar
Mohamed Ali Kaafar