Detecting Privacy Threats with Machine Learning - A Design Framework for Identifying Side-Channel Risks of Illegitimate User Profiling

Aug 4, 2026

Privacy leakage is becoming increasingly severe as IoT, AI, and blockchain deployments expand and generate data-intensive interactions. Such systems can be exploited through side-channel attacks, where adversaries extract sensitive information from devices without directly manipulating the target. The study applies a design science approach to establish a foundation for systematically assessing privacy risks from side-channels. It identifies privacy threats, proposes a machine-learning-driven design framework for detecting side-channel risks, and supports privacy analytics research using keystroke-timing text classification as a use case.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

Introduction

Side-channel attacks and privacy leakage

Research goal and design science approach

Association for Information Systems AIS Electronic Library (AISeL) | AMCIS 2023 Proceedings | Information Security and Privacy (SIG SEC) | | --- | --- | | Aug 10th, 12:00 AM
Detecting Privacy Threats with Machine Learning: A Design Framework for Identifying Side-Channel Risks of Illegitimate User Profiling
Raja Hasnain Anwar
University of Arizona, [rajahasnainanwar@arizona.edu](rajahasnainanwar@arizona.edu)
Yi (Zoe) Zou
University of Western Ontario, [yzou@ivey.ca](yzou@ivey.ca)
Muhammad Taqi Raza
University of Arizona, [taqi@arizona.edu](taqi@arizona.edu)
Follow this and additional works at: [https://aisel.aisnet.org/amcis2023](https://aisel.aisnet.org/amcis2023) | | Recommended Citation Anwar, Raja Hasnain; Zou, Yi (Zoe); and Raza, Muhammad Taqi, "Detecting Privacy Threats with Machine Learning: A Design Framework for Identifying Side-Channel Risks of Illegitimate User Profiling" (2023) . AMCIS 2023 Proceedings. 7. [https://aisel.aisnet.org/amcis2023/sig_sec/sig_sec/7](https://aisel.aisnet.org/amcis2023/sig_sec/sig_sec/7) This material is brought to you by the Americas Conference on Information Systems (AMCIS) at AIS Electronic Library (AISeL) . It has been accepted for inclusion in AMCIS 2023 Proceedings by an authorized administrator of AIS Electronic Library (AISeL) . For more information, please [contact elibrary@aisnet.org](contact elibrary@aisnet.org). Detecting Privacy Threats with Machine Learning: A Design Framework for Identifying Side-Channel Risks of Illegitimate User Profiling Emergent Research Forum (ERF) Paper Raja Hasnain Anwar University of Arizona [rajahasnainanwar@arizona.edu](rajahasnainanwar@arizona.edu) Yi (Zoe) Zou University of Western Ontario [yzou@ivey.ca](yzou@ivey.ca) Muhammad Taqi Raza University of Arizona [taqi@arizona.edu](taqi@arizona.edu) Abstract Privacy leakage has become prevalent and severe with the increasing adoption of the internet of things (IoT), artificial intelligence (AI), and blockchain technologies. Such data-intensive systems are vulnerable to side-channel attacks in which hackers can extract sensitive information from a digital device without actively manipulating the target system. Nevertheless, there is a scarcity of IS research on how businesses can effectively detect and safeguard against side-channel attacks. This study adopts the design science paradigm and lays the groundwork for systematic inquiry into the assessment of privacy risks related to side-channels. In this paper, we a) highlight the privacy threats posed by side-channel attacks, b) propose a machine learning-driven design framework to identify side-channel privacy risks, and c) contribute to the literature on privacy analytics using machine learning techniques. We demonstrate a use case of the proposed framework with a text classification model that uses keystroke timings as side-channel. Keywords Design science, privacy analytics, side-channel attacks, machine learning. Introduction With the ubiquitous use of AI-related computers, smart devices, and sensors, there are increasing concerns among IS professionals and scholars about issues surrounding cybersecurity and data privacy (Rai, 2017). Side-channel attacks have emerged as an increasing threat faced by both business owners and individual consumers. Distinct from other forms of cybersecurity threats where the integrity of an information system is directly compromised through unauthorized system access or malicious system tampering, side-channel attacks are non-invasive and passive. Standaert (2010) defines side-channel attacks as a way for adversaries to learn the physical specifications of the system and the characteristics of its users through externally observable phenomena such as timing information and power consumption. Therefore, side-channel adversaries try to identify unintended information leakage from the activities of connected IS devices. For example, the ZombieLoad attack d

FAQs

What privacy risks do the paper focus on?

The paper focuses on privacy leakage caused by side-channel attacks, where attackers can extract sensitive information without directly manipulating the target system.

How does the proposed framework detect side-channel risks?

It uses a machine learning-driven design framework to identify side-channel privacy risks by guiding data gathering, preprocessing, modeling, and evaluation.

What side-channel signal is used in the demonstrated use case?

The use case applies the framework to user profiling using keystroke timings as the side-channel signal for a text classification model.

Aug 4, 2026

Upload by Lucas Martin