Data Science Group @BCAM | Human-Centered Machine Learning: from Users to Algorithms

​Abstract

Recent advances in machine learning (ML) and artificial intelligence (AI) have created unprecedented opportunities to build interactive systems that support human decisions and behaviors. Yet, ML and AI are often developed with a technology-first mindset, leaving alignment with human needs to be addressed later, during system design. In this talk, I argue that understanding human goals, priorities, and constraints should instead shape the design of ML and AI algorithms from the outset. Achieving this goal requires close collaboration between researchers in human-computer interaction (HCI) and machine learning from the earliest stages of system development. I will illustrate this perspective through a series of projects that integrate ML-driven inference with interactive technologies to support personal health and chronic disease self-management. I will highlight how insights from HCI can shape not only the design of intelligent systems but also the algorithms that power them.