Eric D. Ragan

Human-in-the-Loop User Feedback Affects Perceived Accuracy and Trust, but Task Subjectivity Matters

Donald R. Honeycutt, Mahsan Nourani, Eric D. Ragan

While ML can produce complex models beyond those that a human could produce manually, incorporating human input can often improve performance beyond purely data-driven models. While this feedback could come from system designers or domain experts, in many cases, the end users who…Read more