Associate Professor of Information Systems Beedie School of Business, Simon Fraser University
I study how online reviews shape patients’ choice of doctors, why fake reviews can seem more trustworthy than genuine ones, and how people use AI to make decisions.
I am also a co-founder of Ednius, AI grading software used by universities and certification bodies.
Drawing on a dataset of 34,997 physician reviews, 8,313 of them verified as fake, we find that fake reviews are perceived as more helpful and more trustworthy than genuine ones. They contain the personal and specific health details that real patients, mindful of their privacy, tend to withhold.
Read the paper In the news: SFU News, Medical Economics, HealthDay, PhillyVoice
In a two-stage field experiment on a major healthcare platform, an open appeal raised the likelihood of a patient writing a review by 31% relative to an appeal to help other patients. Appeals that stressed benefits to the reviewer led more patients to reveal sensitive medical information and their identity, while offering privacy control before the review increased identity revelation without reducing what patients shared.
Generative AI makes human-like fake reviews cheap to produce at scale, and existing platform defences have been largely ineffective. This article proposes digital identity verification, in which a platform confirms who the reviewer is while the published review remains anonymous.
Clickstream data from a doctor appointment booking platform show that where ratings are abundant, patients consider fewer doctors, browse for less time, and focus on doctors closer to home. Where ratings are scarce, they search longer and farther afield. Highly rated doctors gain demand at the expense of unrated ones.
Students often wait weeks for feedback that is too generic to act on, and educators spend 10 to 22 hours a week grading per course. Ednius, which I co-founded in 2024, is an AI-assisted grading platform for universities and certification bodies. It drafts personalized, rubric-aligned feedback on open-ended student work, and the instructor retains final approval on every grade.
Ednius is built the way I run a study: every feature ships behind an A/B test, and design decisions follow field-experiment evidence. It is also a live setting for research on human–AI decision-making, and I welcome collaborators.
How the AI grading software works → The Ednius story and call for research collaboration →
I am an Associate Professor of Information Systems at the Beedie School of Business, Simon Fraser University. My training spans technical foundations (a bachelor’s in Computer Science and Engineering), management (an MBA), and research (a PhD in Information Systems). That mix shapes how I work across disciplines.
My research examines digital word-of-mouth in healthcare, a setting where the stakes of a choice are high and the quality of a service is hard to judge. I combine field experiments, natural experiments, and machine learning (including large language models) on data from live platforms, with the aim of producing evidence that platform designers and policymakers can act on.
Before my academic career, I worked in the technology sector, with roles at Tata Motors and Practo. That exposure to how IT actually gets planned, built, and adopted inside organizations still shapes how I think about the practical side of my research.