Aminul Islam

PhD Candidate in Computer Science

University of Illinois Chicago

Contact: mislam34@uic.edu

About

I am a PhD candidate in Computer Science at the University of Illinois Chicago, advised by Professor Elena Zheleva.

My research interests broadly lie at the intersection of data science and machine learning, with a focus in personalized recommendation, ranking, and search. My current research focuses on recommender systems by addressing biases such as position, exposure, and popularity bias. I work across learning-to-rank, collaborative filtering, graph-based personalized recommender systems, RecLLMs, and social recommender systems, using methods from causal inference, econometrics, and machine learning to mitigate these biases. My work also includes unbiased evaluation of recommender systems. More broadly, I am interested in compelling real-world applications within recommender systems, e-commerce, and data science. I also work on causal hypothesis verification using LLMs and mentor undergraduate students on this project.

The broader impact of my research spans several real-world platforms, including e-commerce, social networks, video-sharing platforms, lifestyle applications, search engines, and many more. I am broadly interested in other compelling applications within recommender systems, e-commerce, and broader data science related works.

Education

PhD Candidate in Computer Science (Jan. 2023 to present)

University of Illinois Chicago
M.Sc in Computer Science

University of Illinois Chicago
B.Sc in Computer Science and Engineering

Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh

Work Experience

Graduate Research Assistant, University of Illinois Chicago May 2024 to present

I am working with Professor Elena Zheleva. I am involved in projects that address various biases in learning-to-rank systems, graph-based recommender systems, and social recommender systems.

Internship, IDEAL Research Institute May 2025 to Oct 2025

I have worked as a visiting student researcher at IDEAL (The Institute for Data, Econometrics, Algorithms, and Learning) with Professor Ren Wang. I have worked on addressing popularity bias in GNN-based collaborative filtering in a post-hoc manner. Paper accepted at ACM Web Conference 2026.

Graduate Teaching Assistant, University of Illinois Chicago Jan. 2023 to May 2024 and Jan. 2026 to May 2026

I worked as a teaching assistant for Introduction to Data Science (CS 418) and Programming Practicum (CS 211) courses.

Software Engineer, Ridmik Labs, Dhaka, Bangladesh Nov. 2018 to Dec. 2022

I was a Software Engineer at Ridmik Labs, where I was one of the primary developers of the Bengali Ridmik Keyboard, which has over 200 million users. I also worked on large-scale native mobile apps such as Boitoi-Android and Boitoi-iOS, implementing most features and functionalities development. Additionally, I worked developing a Bengali news article summarizer, web applications, backend systems, REST APIs, end-to-end Web2 encryption, account kits, and security features in software applications.

AIST Web Application, Department of CS, BUET May 1, 2021 to July 31, 2021

AIST Project is a web application for visualizing the predictions of AIST (An Interpretable Attention-based Deep Learning Model for Crime Prediction).

Publications

[1] Md Aminul Islam, Elena Zheleva, Ren Wang. Post-hoc Popularity Bias Correction in GNN-based Collaborative Filtering. In Proceedings of the ACM Web Conference 2026. PDF

[2] Md Aminul Islam, Kathryn Vasilaky, Elena Zheleva. A Control Function Framework for Mitigating Position Bias in Learning to Rank Systems. In Proceedings of the ACM Conference on Recommender Systems 2026. PDF

[3] Md Aminul Islam. Unbiased Recommender Systems with Implicit Feedback. In Proceedings of the ACM Conference on Recommender Systems 2026 (Doctoral Symposium). PDF

[4] Safiyyah Ahmed, Abrar Ansari, Md Aminul Islam, Elena Zheleva. Medical Causal Hypothesis Verification with Large Language Models. CONSEQUENCES (Causality, Counterfactuals and Sequential Decision-Making for Recommender Systems) Workshop @ RecSys 2026. PDF

[5] Md Aminul Islam, Ahmed Sayeed Faruk, Sourav Medya, Elena Zheleva. Debiasing Message Passing to Mitigate Popularity Bias in GNN-based Collaborative Filtering. [arXiv]

Other Research Projects

[1] Md Aminul Islam, Md Mezbaur Rahman, Mehrab Mustafy Rahman. Can Large Language Models Mitigate Popularity Bias in Personalized Recommender Systems? PDF

[2] Md Aminul Islam. Robust Learning-to-Rank Against Noisy Clicks with Residual-Based Correction. PDF

[3] Md Aminul Islam, Ahmed Sayeed Faruk. Prompt-Based LLMs for Position Bias-Aware Reranking in Personalized Recommendations. [arXiv]

[4] Md Aminul Islam. Double Machine Learning for Selection Bias Recovery in Learning-to-Rank Systems. PDF

Skills

Programming Languages: Python, Java, Swift, Kotlin, C, C++, Assembly (80X86)
Scripting Languages: JavaScript, Shell Scripts, Java Native Interface (JNI), HTML, CSS
Mobile Application Development & Frameworks: Android, iOS, Django
Database: SQL, Oracle
Security: AES (CBC & ECB), CommonCrypto API, iOS Keychain
Tools & Others: Pandas, PyTorch, Tensorflow, LLM APIs, Google Cloud, REST API, Firebase, Google ML Kit & Vision API, LaTeX