I gave this talk on November 19, 2021 for Vector Institute.
Every day, information access systems mediate our experience of the world beyond our immediate senses. Google and Bing help us find what we seek, Amazon and Netflix recommend things for us to buy and watch, Apple News gives us the day’s events, and BuzzFeed guides us to related articles. These systems deliver immense value, but also have profound influence on how we experience information and the resources and perspectives we see. There are significant challenges, however, in measuring this influence and characterizing the benefits and harms these systems deliver to the various people they affect.
In this talk, I will present our work on the question “what does it mean for an information access system to be good for people?”. Through a combination of system-building, experimentation, and data analysis, my collaborators and I are working to provide some answers to this question. I will report on several projects on understanding information needs, quantifying systematic biases in recommender system outputs and evaluation, and discuss what it takes to make recommendation, retrieval, and the other algorithmic components of information work for people.
These papers provide more details on the research I presented. Many of them have accompanying code to reproduce the experiments and results.