An international research team developed a visual analytics tool to address challenges faced by child welfare specialists using machine learning models to screen cases.
Researchers at the Massachusetts Institute of Technology (MIT), New Zealand's Auckland University of Technology, and Australia's University of Queensland used feedback from call screeners at a child welfare department in Colorado in designing the tool.
With the help of bar graphs, the tool shows how specific factors of a case contribute to the predicted risk that a child will be removed from their home within two years.
MIT's Kalyan Veeramachaneni said, "A big takeaway from this project is that these domain experts don't necessarily want to learn what machine learning actually does. They are more interested in understanding why the model is making a different prediction than what their intuition is saying, or what factors it is using to make this prediction."
From MIT News
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA