Affiliation:
1. University of Notre Dame, USA
2. Hong Kong University of Science and Technology, China
3. Microsoft Research, USA
Abstract
Fairness measurement is crucial for assessing algorithmic bias in various types of machine learning (ML) models, including ones used for search relevance, recommendation, personalization, talent analytics, and natural language processing. However, the fairness measurement paradigm is currently dominated by fairness metrics that examine disparities in allocation and/or prediction error as univariate key performance indicators (KPIs) for a protected attribute or group. Although important and effective in assessing ML bias in certain contexts such as recidivism, existing metrics don’t work well in many real-world applications of ML characterized by imperfect models applied to an array of instances encompassing a multivariate mixture of protected attributes, that are part of a broader process pipeline. Consequently, the upstream representational harm quantified by existing metrics based on how the model represents protected groups doesn’t necessarily relate to allocational harm in the application of such models in downstream policy/decision contexts. We propose FAIR-Frame, a model-based framework for parsimoniously modeling fairness across multiple protected attributes in regard to the representational and allocational harm associated with the upstream design/development and downstream usage of ML models. We evaluate the efficacy of our proposed framework on two testbeds pertaining to text classification using pretrained language models. The upstream testbeds encompass over fifty thousand documents associated with twenty-eight thousand users, seven protected attributes and five different classification tasks. The downstream testbeds span three policy outcomes and over 5.41 million total observations. Results in comparison with several existing metrics show that the upstream representational harm measures produced by FAIR-Frame and other metrics are significantly different from one another, and that FAIR-Frame’s representational fairness measures have the highest percentage alignment and lowest error with allocational harm observed in downstream applications. Our findings have important implications for various ML contexts, including information retrieval, user modeling, digital platforms, and text classification, where responsible and trustworthy AI are becoming an imperative.
Publisher
Association for Computing Machinery (ACM)
Subject
Computer Science Applications,General Business, Management and Accounting,Information Systems
Reference108 articles.
1. Sentiment analysis in multiple languages: Feature selection for opinion classification in web forums;Abbasi Ahmed;ACM Transactions on Information Systems (TOIS),2008
2. Data Science for Social Good;Abbasi Ahmed;Journal of the Association for Information Systems,2023
3. Ahmed Abbasi, David Dobolyi, John P Lalor, Richard G Netemeyer, Kendall Smith, and Yi Yang. 2021. Constructing a psychometric testbed for fair natural language processing. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 3748–3758.
4. Selecting attributes for sentiment classification using feature relation networks;Abbasi Ahmed;IEEE Transactions on Knowledge and Data Engineering,2010
5. Ahmed Abbasi, Jingjing Li, Gari Clifford, and Herman Taylor. 2018. Make “fairness by design” part of machine learning. Harvard Business Review(2018).