Bamshad Mobasher

Bamshad Mobasher

Professor of Computer Science @ DePaul  |  Founding Director, DePaul AI Institute  |  Researching personalization, recommender systems, and the responsible use of AI

Welcome! I'm a Professor of Computer Science in the Jarvis College of Computing and Digital Media at DePaul University in Chicago. I'm the founding director of the DePaul AI Institute, I direct the Center for Web Intelligence, and I chair the School of Computing's Artificial Intelligence program.

I earned my Ph.D. in Computer Science from Iowa State University in 1994 and before joining DePaul I was on the faculty at the University of Minnesota.

My work lives at the intersection of artificial intelligence, machine learning, and user modeling, and I'm fortunate to be recognized as one of the leading authorities on algorithmic personalization and recommender systems. Over my career I've published five books and more than 300 scientific articles; my work has been cited over 34,000 times.

Beyond academia, I've worked with or advised many companies such as IBM, Amazon, Pandora, TiVo, and Meta on personalization and predictive modeling. I have also helped lead the field's flagship venues, including ACM RecSys (as founding member and steering committee chair) and ACM UMAP (as a director on the conference series Board), and serve as an associate editor for several ACM Transactions journals, including the ACM Transactions on Recommender Systems and the ACM Transactions on Intelligent Interactive Systems.

Research

My research centers on how intelligent systems learn from human behavior and shape what people see. Early on, I helped lay the foundations of web data mining and automatic personalization. Today, my students and I focus increasingly on the fairness, bias, and user agency of the recommendation algorithms now embedded in everyday life; on the integration of context-awareness into intelligent personalization algorithms; and on how large language models can model evolving user preferences.

# Research Areas

My research spans the systems that learn from human behavior and shape what people encounter — from the algorithms behind personalization to the fairness and trustworthiness of those algorithms at scale.

01

Recommender Systems & Personalization

Designing and evaluating algorithms that tailor content, products, and information to individual users. Key problems: balancing accuracy against diversity and novelty; building context-aware and session-based recommenders that adapt to a user's situation and intent; and, most recently, using large language models to reason about preferences and generate recommendations that are both effective and explainable.

02

Fairness, Bias & User Agency

Studying how recommendation algorithms distribute attention and opportunity, and how that shapes what people see. Key problems: measuring and mitigating popularity bias and multi-sided (multistakeholder) exposure bias; understanding the feedback loops that amplify bias over time; and giving users meaningful transparency, control, and agency over the systems that filter their world.

03

Predictive User Modeling

Building models of users that capture not only static taste but how preferences shift across time and context. Key problems: detecting changes in user intent with sequential and session-aware models; modeling temporal dynamics using hidden Markov models and, increasingly, LLM-based profiling; and inferring rich context to anticipate what a user will need next.

04

Trustworthy & Secure Recommendation

Examining how robust personalization systems are when adversaries try to manipulate them. Key problems: characterizing profile-injection ("shilling") attacks against collaborative filtering; detecting obfuscated and segment-focused attacks; and designing model-based algorithms and anomaly-detection methods that keep recommendations trustworthy under attack.

05

AI Ethics & Intelligent Agents

Investigating the societal, ethical, and legal dimensions of AI alongside the design of autonomous agents. Key problems: algorithmic knowledge manipulation and misinformation (what I call algorithmic "agnotology"); the responsible governance of agentic AI; and multi-agent systems for negotiation, automated contracting, and electronic commerce.

06

Web and Social Media Mining

Discovering patterns from the traces people leave across the web and social platforms — the foundational area where my early work helped establish the field of web data mining. Key problems: mining usage, content, structure, and social-tagging data; reconstructing sessions and aggregate usage profiles; and integrating semantic and social knowledge to power personalization at web scale.

# The DePaul AI Institute

Since 2024 I've served as the founding director of the DePaul AI Institute.

Founding Director · 2024–present

An interdisciplinary home for AI at DePaul

The Institute brings together faculty from across the university — computing, healthcare, law, business, the natural and social sciences, and the arts — to advance research and education on both the promises and the challenges of AI, with a steady commitment to applying it for the greater social good.

AI Ethics & SocietyAI in BusinessAI in Healthcare & ScienceAI in the ArtsAI in TeachingAI for Social Good
Visit the Institute → Our mission AI-Institute@depaul.edu

# Appointments & Experience

# News

2025The DePaul AI Institute hosted a series of symposia and conferences across the year — including the Symposium on AI in the Arts, the Symposium on AI in Business, and the Symposium on AI in Healthcare and Scientific Discovery.
2024Our NSF project "AI-Enabled Support Services for Migrants" (ReDDDoT) began — a community-centered design effort running through 2026.
2024I became the founding director of the DePaul AI Institute, a new interdisciplinary center for AI research, education, and ethics.
2024I continue to co-organize the Workshop on Context-Aware Recommender Systems at ACM RecSys.

Recently Accepted & Published

  1. Towards Explainable Temporal User Profiling with LLMs. M. Sabouri, M. Mansoury, K. Lin, B. Mobasher. Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization (UMAP '25), New York, July 2025.
  2. "They Only Offer the Illusion of Choice": Exploring User Perceptions of Control and Agency on YouTube. M. F. Ghori, A. Dehpanah, J. Gemmell, B. Mobasher. Adjunct Proceedings of the 33rd ACM UMAP, New York, July 2025.
  3. Using LLMs to Capture Users' Temporal Context for Recommendation. M. Sabouri, M. Mansoury, K. Lin, B. Mobasher. Workshop on Context-Aware Recommender Systems, ACM RecSys, Prague, September 2025.
  4. A Hybrid Recommendation Framework for Enhancing User Engagement in Local News. P. Pourashraf, B. Mobasher. Proceedings of HCI International, Gothenburg, Sweden, June 2025.
  5. Mitigating Exposure Bias in Online Learning to Rank Recommendation: A Novel Reward Model for Cascading Bandits. M. Mansoury, B. Mobasher, H. van Hoof. Proceedings of the ACM Conference on Information and Knowledge Management (CIKM '24), 2024.
  6. Beyond Static Calibration: The Impact of User Preference Dynamics on Calibrated Recommendation. K. Lin, M. Mansoury, F. Eskandanian, M. Sabouri, B. Mobasher. Proceedings of the 32nd ACM UMAP, Sardinia, Italy, 2024.

# Recent Talks

A selection of recent keynotes, invited talks, and panels.

Recorded talks & videos. The DePaul AI Institute maintains a growing library of recorded presentations from its symposium and "AI in Teaching" series, where faculty from across the university demonstrate research and classroom uses of AI.

Watch recorded talks at the AI Institute →

# Publications

Drawn from more than 300 articles. Below are recent highlights and a few foundational papers; the complete, up-to-date record lives on Google Scholar and DBLP.

Selected Recent Publications

  1. UMAP
    2025
    Towards Explainable Temporal User Profiling with LLMs
    Milad Sabouri, Masoud Mansoury, Kun Lin, Bamshad Mobasher
    33rd ACM Conference on User Modeling, Adaptation and Personalization
  2. UMAP
    2025
    "They Only Offer the Illusion of Choice": Exploring User Perceptions of Control and Agency on YouTube
    Muheeb Faizan Ghori, Arman Dehpanah, Jonathan Gemmell, Bamshad Mobasher
    ACM UMAP (Adjunct)
  3. CIKM
    2024
    Mitigating Exposure Bias in Online Learning to Rank: A Reward Model for Cascading Bandits
    Masoud Mansoury, Bamshad Mobasher, Herke van Hoof
    ACM Conference on Information and Knowledge Management
  4. UMAP
    2024
    Beyond Static Calibration: The Impact of User Preference Dynamics on Calibrated Recommendation
    Kun Lin, Masoud Mansoury, Farzad Eskandanian, Milad Sabouri, Bamshad Mobasher
    32nd ACM UMAP
  5. ACM TOIS
    2021
    A Graph-Based Approach for Mitigating Multi-Sided Exposure Bias in Recommender Systems
    Masoud Mansoury, Himan Abdollahpouri, M. Pechenizkiy, Bamshad Mobasher, Robin Burke
    ACM Transactions on Information Systems, 40(2)
  6. RecSys
    2020
    The Connection Between Popularity Bias, Calibration, and Fairness in Recommendation
    Himan Abdollahpouri, Masoud Mansoury, Robin Burke, Bamshad Mobasher
    14th ACM Conference on Recommender Systems

Featured Earlier Publications

  1. Automatic Personalization Based on Web Usage Mining. B. Mobasher, R. Cooley, J. Srivastava. Communications of the ACM, 43(8), 142–151, 2000. — one of the field's most-cited papers (1,500+ citations).
  2. Data Preparation for Mining World Wide Web Browsing Patterns. R. Cooley, B. Mobasher, J. Srivastava. Knowledge and Information Systems, 1(1), 5–32, 1999.
  3. Discovery and Evaluation of Aggregate Usage Profiles for Web Personalization. B. Mobasher, H. Dai, T. Luo, M. Nakagawa. Data Mining and Knowledge Discovery, 6(1), 61–82, 2002.
  4. Toward Trustworthy Recommender Systems: An Analysis of Attack Models and Algorithms. B. Mobasher, R. Burke, C. Williams, R. Bhaumik. ACM Transactions on Internet Technology, 7(4), 2007.
  5. Context-Aware Recommender Systems. G. Adomavicius, B. Mobasher, F. Ricci, A. Tuzhilin. AI Magazine, 32(3), 67–80, 2011.

Books

  • Intelligent Techniques for Web Personalization — Mobasher & Anand (eds.), Springer LNAI 3169, 2005
  • Advances in Web Mining and Web Usage Analysis, Vols. I–IV — Springer LNCS, 2006–2009
Google Scholar ↗ DBLP ↗ Semantic Scholar ↗

# Teaching

I teach graduate and undergraduate courses in artificial intelligence, machine learning, data mining, and intelligent systems.

# Professional Service

A selection of featured editorial roles and conference leadership; this is a representative list rather than a complete one.

Featured Editorial Roles

Selected Conference & University Leadership

  • Steering Committee Chair & founding member, ACM RecSys (chair 2020–2023)
  • Treasurer & Director, User Modeling Inc. — steering body for ACM UMAP
  • General Co-Chair, ACM RecSys 2011 · Program Co-Chair, ACM UMAP 2012
  • Co-founder, the WebKDD workshop series — helped establish the field of web data mining
  • President, Faculty Council, DePaul University (2016–2018)

# Contact

I'm glad to talk about research, teaching, the AI Institute, or potential collaborations.

Email
mobasher@cs.depaul.edu
Office Phone
(312) 362-5174
Office
CDM Center, Room 833
Department
School of Computing, Jarvis College of CDM
Address
243 South Wabash Ave, Chicago, IL 60604
Faculty page
facweb.cs.depaul.edu/mobasher