Marian-Andrei Rizoiu

Professional Webpage

Marian-Andrei Rizoiu.

I am a Professor leading the Behavioral Data Science lab at the University of Technology Sydney, and Director of the Defence Innovation Network (DIN). My interdisciplinary research crosses computer and social sciences, blending psycholinguistics, digital communication and stochastic modelling to understand human attention dynamics in the online environment, the emergence of influence and opinion polarisation. I am a two-time finalist for the prestigious Australian Museum Eureka Prize for Outstanding Science in Safeguarding Australia (2024, 2025), and recipient of the Excellence Award and Academic of the Year at the 2023 Australian Defence Industry Awards. I currently lead grants worth $11.8 million from the Commonwealth of Australia, including $8.56 million from Defence's Advanced Strategic Capabilities Accelerator (ASCA), to detect and model the spread of mis- and disinformation and its weaponised counterparts – information and influence operations.

Research

My research has made several key contributions to information operations detection, online misinformation prediction, and understanding the radicalisation pathways in online environments.

First, I have developed paradigm-shifting behavioural models for detecting state-sponsored information operations. My approach focuses on behavioural fingerprinting rather than content analysis, enabling detection that cannot be easily bypassed by adversaries. The IC-Mamba system predicts misinformation virality before it occurs, whilst the X-Troll framework provides explainable detection of state-sponsored agents using appraisal theory and linguistic analysis. Second, I have built early warning systems and analytical dashboards (NARRATE) that are deployed for operational use by defence and intelligence agencies. These technologies provide asymmetric advantages through real-time detection, attribution, and decision support capabilities.

See more about my research.

News

See the lab's news page for more recent news.

2026-08: I presented "Behavioural fingerprinting for detection and triage of covert IO agents" at ADSTAR 2026 in Adelaide, alongside Lin Tian's talk on hardening information-operation detectors against adaptive adversaries. I also joined the "ADSUN as a National System" panel as Director of the Defence Innovation Network.

2026-07: I was promoted to Professor at the University of Technology Sydney.

2026-01: Our paper "DREAMS: A Social Exchange Theory-Informed Modeling of Misinformation Engagement on Social Media" by Lin Tian and Marian-Andrei Rizoiu was accepted at The Web Conference 2026 (WWW'26) in Dubai! Acceptance rate: 20.1%.

2025-09: Appointed Director of the Defence Innovation Network (DIN), a network of 9 NSW and ACT universities facilitating Defence capability transfer into the NSW industrial ecosystem, managing a $2.3M annual budget.

Back to Top

Research topics.

My research asks how harmful information spreads online, and how it can be detected and countered before it causes real-world harm. The central move is a shift from content-based to behaviour-based detection. Rather than analysing what is said — which raises censorship concerns, requires language-specific tools, and is easily evaded by an adversary who simply changes wording — my methods analyse how information spreads: the behavioural signatures and propagation patterns that coordinated actors cannot hide without abandoning their objective. This makes detection language-agnostic and adversary-resistant, and it is what allows the methods to transfer across platforms and languages.

A second finding, arising from work funded by the Department of Home Affairs, is that misinformation is a problem of social belonging rather than a problem of information. Debunking a falsehood rarely changes minds, because people filter what they read through pre-existing worldviews and community allegiances. This has shifted policy approaches away from fact-checking alone and towards community-level intervention.

I also work on skills and labour markets. Using large collections of online job advertisements, I quantify how similar skills are to one another and build occupation-transition recommenders: which skills a worker can carry into a new occupation, which occupations suit a given set of skills, and where the training gaps lie. The same methods measure how effectively skills are deployed across a workforce.

Where this is heading

The through-line of my work is that online harm is a systems problem rather than a content problem. Whether the subject is a state-sponsored influence campaign, a misinformation cascade during a natural disaster, or a radicalisation pathway, the useful signal lies in how people and populations behave around information, not in the words themselves. That framing is what makes the research portable: a method built for one language and one platform should carry to the next, because it never depended on either to begin with.

Two commitments follow. The first is that detection is only half the problem. Knowing something is spreading is worth little without knowing how much time remains to act and which intervention would help, so my work aims increasingly at forecasting and at the design of interventions rather than at classification alone. The second is that these systems have to be explainable. An analyst or a policymaker cannot act on a score; they need to know why an account, a narrative or a community was flagged, and to be able to disagree with the answer.

I work at the boundary between computer science and the social sciences, which is where I think the interesting problems now sit. The mathematics is only useful insofar as it is disciplined by how people actually behave. My research is built and tested with government and defence partners, and it is meant to end up in the hands of the people who have to make decisions with it.

Back to Top

Grants, funding and projects.

Funding

I have secured $12.1 million in research funding, including $8.56 million from Defence's Advanced Strategic Capabilities Accelerator across three major projects. The work is supported by Defence and national-security agencies — ASCA, the Defence Science and Technology Group, the Office of National Intelligence and the Department of Home Affairs — alongside the Australian Research Council, state government, industry and philanthropy.

The portfolio was built in stages, each designed to earn the next: early contracts to establish the theory and the algorithms; operational prototyping with national agencies; demonstration at regional scale in the Pacific; and now large multi-institution programmes that put the methods in front of the people who have to use them.

Selected funded research

Other funded research

Where the work is going

Most of my current portfolio concentrates on detecting and countering information operations and online misinformation, delivered with Defence, government and university partners in Australia and internationally. A second strand applies the same behavioural methods to skills and labour markets, supporting government workforce and training policy. A third and newer strand examines safety and rule-evasion in multi-agent AI systems.

Alongside the research I direct the Defence Innovation Network, which connects defence research across nine universities in New South Wales and the ACT and channels university capability into the defence industrial base.

Current projects, publications and software releases are listed on the Behavioral Data Science lab site.

Back to Top

Supervision and teaching.

Research supervision

Most of my teaching now happens through research supervision. I lead a group of postdoctoral researchers and PhD students in the Behavioral Data Science lab, working across computer science and the social sciences.

My approach is to give students problems that matter to someone outside the university, and the room to make them their own.

Prospective students

The lab takes students from machine learning, applied statistics and computational social science, as well as those arriving from the social sciences who want to work quantitatively.

Current openings, expectations and lab details are kept on the lab site.

Earlier teaching

I have previously convened and lectured undergraduate and postgraduate courses at the Australian National University and at Lumière University Lyon, covering relational databases and data warehousing, document analysis, information retrieval and natural language processing, and numerical machine learning.

Back to Top

Biography.

I am a Professor in Computer Science with the Faculty of Engineering and IT in the University of Technology Sydney, and Director of the Defence Innovation Network.

At UTS

I joined UTS in February 2019 and have held each academic level here since.

  • September 2025 – present — Director, Defence Innovation Network
  • July 2026 – present — Professor, Faculty of Engineering and IT
  • January 2024 – July 2026 — Associate Professor in Behavioral Data Science, Data Science Institute
  • July 2021 – January 2024 — Senior Lecturer in Behavioral Data Science, Data Science Institute
  • February 2019 – July 2021 — Lecturer in Computer Science, Faculty of Engineering and IT

Previously

Between March 2016 and January 2019, I was a Research Fellow, then Lecturer with the College of Engineering and Computer Science at the Australian National University in Canberra. I was equally affiliated with the Data61 unit of CSIRO, in the Decision Sciences team.

Between May 2014 and February 2016,I was a researcher within the National ICT Australia in Canberra Australia, working in the Optimization Research Group. I was equally an adjunct lecturer with the College of Engineering and Computer Science at the Australian National University in Canberra.

Between September 2013 and May 2014, I was a PostDoctoral researcher with the ERIC Laboratory, financed by the ImagiWeb Research Project. I was equally an assistant professor with the DIS Department, at the University Lumière in Lyon.

Between 2009 and 2013, I was a PhD student at the ERIC Laboratory with the University Lumière Lyon2, under the supervision of Stéphane Lallich and Julien Velcin. I defended my PhD thesis on June 24th 2013, with honors "Très Honorable".

In July 2009, I obtained my MSc (graduating first of promotion, with honors.) in Data Mining and Knowledge Management from the Polytechnic School of the University of Nantes, France and wrote my Master’s Thesis on "Textual Data Clustering and Cluster Naming" after an internship at the ERIC Laboratory.

Between 2004 and 2009, I did my undergrad and obtained in September 2009 my Engineer Diploma (double diploma, in parallel with the French Master's) in System and Computer Engineering from the Faculty of Automatic Control and Computers of the Polytechnic University Bucharest, Romania.

Back to Top

Publications.

A selection of the work I consider most significant, rather than the most recent. The complete list is maintained on the Behavioral Data Science lab publications page.

Information operations and misinformation

Modelling information diffusion

Skills, labour markets and people

Research student theses

Back to Top