Metrics for Dataset Demographic Bias: A Case Study on Facial Expression Recognition
This work addresses dataset bias for researchers in AI to improve fairness and accuracy, but it is incremental as it focuses on metric classification and redundancy analysis rather than introducing new methods.
The authors tackled the problem of measuring demographic bias in datasets, particularly in facial expression recognition, by reviewing and classifying existing metrics and conducting a case study on 20 datasets, finding that many metrics are redundant and a reduced subset may suffice for bias measurement.
Demographic biases in source datasets have been shown as one of the causes of unfairness and discrimination in the predictions of Machine Learning models. One of the most prominent types of demographic bias are statistical imbalances in the representation of demographic groups in the datasets. In this paper, we study the measurement of these biases by reviewing the existing metrics, including those that can be borrowed from other disciplines. We develop a taxonomy for the classification of these metrics, providing a practical guide for the selection of appropriate metrics. To illustrate the utility of our framework, and to further understand the practical characteristics of the metrics, we conduct a case study of 20 datasets used in Facial Emotion Recognition (FER), analyzing the biases present in them. Our experimental results show that many metrics are redundant and that a reduced subset of metrics may be sufficient to measure the amount of demographic bias. The paper provides valuable insights for researchers in AI and related fields to mitigate dataset bias and improve the fairness and accuracy of AI models. The code is available at https://github.com/irisdominguez/dataset_bias_metrics.