Login / Signup

Investigation of bias in the automated assessment of school violence.

Lara J KanbarAnagh MishraAlexander OsbornAndrew CifuentesJennifer CombsMichael SorterDrew BarzmanJudith W Dexheimer
Published in: Journal of biomedical informatics (2024)
Bias assessment is needed to address shortcomings within machine learning. In our work, student race, ethnicity, sex, use of public assistance, and annual household income did not explain ARIA's risk assessment score of students. ARIA will continue to be evaluated regularly with increased subject recruitment.
Keyphrases
  • machine learning
  • mental health
  • risk assessment
  • high school
  • physical activity
  • healthcare
  • deep learning
  • artificial intelligence
  • heavy metals
  • high throughput
  • big data
  • climate change
  • single cell