Fair Machine Learning with Structured and Unstructured Features: A Case Study in Mortality Prediction
- Resource Type
- Conference
- Authors
- Chen, Hanyang; Zhao, Zhouting; James Ng, Tin Lok
- Source
- 2023 31st Irish Conference on Artificial Intelligence and Cognitive Science (AICS) Artificial Intelligence and Cognitive Science (AICS), 2023 31st Irish Conference on. :1-4 Dec, 2023
- Subject
- Computing and Processing
General Topics for Engineers
Robotics and Control Systems
Signal Processing and Analysis
Pipelines
Machine learning
Medical services
Cognitive science
Task analysis
Fairness
Structured and unstructured features
Bias mitigation
Healthcare
- Language
Fairness in machine learning has garnered significant attention in both academic and industrial circles, especially in domains like healthcare. Many predictive problems involve a mix of structured and unstructured features. While existing fairness-aware machine learning methods often focus on either structured or unstructured features, addressing fairness in the presence of both is a relatively unexplored area. Our approach involves developing two distinct machine learning pipelines that incorporate bias mitigation techniques. We apply these methods to a binary classification task of predicting patient mortality during hospitalizations.