Publications

How Information Retrieval Systems Construct and Amplify Immigration Narratives

Published in European Conference on Information Retrieval (ECIR 2026), IR-for-Good Track, 2026

This paper examines how modern information retrieval systems shape, construct, and amplify dominant narratives around immigration. Using large-scale IR datasets and qualitative framing analysis, we show how retrieval pipelines can systematically privilege particular viewpoints, raising concerns for fairness, accountability, and responsible AI.

Partisan perspectives on AI and immigration: An analysis of Canadian parliamentary discourse (2014–2024)

Published in Canadian Ethnic Studies (in press), 2025

This paper analyzes how Canadian parliamentarians frame artificial intelligence and immigration between 2014 and 2024, revealing partisan divides in discourse.

Download Paper

Sinophobia on Twitter: Integrating Theory, Humans, and Machines

Published in ACM CSCW Workshop on Addressing Challenges and Opportunities in Online Extremism Research, 2021

This workshop paper explores Sinophobic discourse on Twitter by integrating social theory, human annotation, and machine learning. It highlights the value of interdisciplinary approaches for studying online extremism and racialized hate.

SAFFRON: A Semi-Automated Framework for Software Requirements Prioritization

Published in International Journal of Advanced Computer Science and Applications, Vol. 8, No. 12, 2017

This paper presents SAFFRON, a semi-automated framework that integrates natural language processing and stakeholder feedback to improve the prioritization of software requirements. It addresses scalability and subjectivity challenges in traditional requirement analysis workflows.

Download Paper

Switching GAN-Based Image Filters to Improve Perception for Autonomous Driving

Published in Master's Thesis, University of Waterloo, 2019

This thesis investigates the use of generative adversarial networks to improve visual perception in autonomous driving systems by dynamically switching image filters under varying environmental conditions.

Download Paper

Switching GAN-Based Image Filters to Improve Perception for Autonomous Driving

Published in University of Waterloo Thesis, 2019

A dissertation exploring GAN-based image enhancement techniques for improved autonomous driving perception.

Download Paper

The Role of Artificial Intelligence in Immigration: From Automation to Accountability

Published in In Ana Beduschi (Ed.), Handbook on Migration and Artificial Intelligence, Edward Elgar Publishing, 2026

This chapter provides a critical overview of how artificial intelligence is deployed across immigration systems, from application processing to border enforcement. It highlights risks related to bias, opacity, and accountability, and outlines pathways for rights-respecting and transparent AI governance in migration contexts.

State-of-the-Art Translation of Text-to-Gloss Using mBART: A Case Study of Bangla

Published in arXiv, 2025

This paper presents a state-of-the-art neural approach for translating Bangla text into gloss representations using mBART. It contributes to low-resource sign language processing and accessibility research.

Download Paper

Stemming: The Evolution and Current State with a Focus on Bangla

Published in arXiv, 2025

This paper surveys the evolution of stemming algorithms with a particular focus on Bangla. It reviews linguistic challenges, existing methods, and open research problems in low-resource language processing.

Download Paper

Where Journalism Silenced Voices: Indigenous Communities in Bangladeshi Media

Published in arXiv, 2025

This study examines how mainstream Bangladeshi journalism marginalizes Indigenous communities through framing, omission, and narrative silencing. It combines computational text analysis with critical media theory.

Download Paper