A Systematic Mapping on Software Fairness - Appendix

APPENDIX I LIST OF RESEARCH STUDIES ABOUT SOFTWARE FAIRNESS Table-A I-1 Research Studies Analyzed in the Systematic Literature Mapping (1-42) # Studies Selected for the Systematic Literature Mapping S1 ’Ignorance and Prejudice’ in software fairness S2 Astraea: Grammar-Based Fairness Testing S3 Bias in machine learning software: Why? how? what to do? S4 Fairness in Design: A Framework for Facilitating Ethical Artificial Intelligence Designs S5 FairMask: Better Fairness via Model-Based Rebalancing of Protected Attributes S6 Fix Fairness, Don’t Ruin Accuracy: Performance Aware Fairness Repair using AutoML S7 Information-Theoretic Testing and Debugging of Fairness Defects in Deep Neural Networks S8 Fairness Improvement with Multiple Protected Attributes: How Far Are We? S9 An Empirical Study on Correlations between Deep Neural Network Fairness and Neuron Coverage Criteria S10 Fair Enough: Searching for Sufficient Measures of Fairness S11 Perfectly parallel fairness certification of neural networks S12 RUNNER: Responsible UNfair NEuron Repair for Enhancing Deep Neural Network Fairness S13 Automatic Fairness Testing of Machine Learning Models S14 Fairway: A way to build fair ML software S15 Fairify: Fairness Verification of Neural Networks S16 MAAT: a novel ensemble approach to addressing fairness and performance bugs for machine learning software S17 MBFair: a model-based verification methodology for detecting violations of individual fairness S18 Metamorphic testing and certified mitigation of fairness violations in NLP models S19 ReFAIR: Toward a Context-Aware Recommender for Fairness Requirements Engineering S20 Fairness testing: Testing software for discrimination S21 Training Data Debugging for the Fairness of Machine Learning Software S22 Automated directed fairness testing S23 Search-based fairness testing for regression-based machine learning systems S24 Search-based Automatic Repair for Fairness and Accuracy in Decision-making Software S25 The Impact of Data Preparation on the Fairness of Software Systems S26 Improving Fairness in Speaker Recognition S27 Identifying Imbalance Thresholds in Input Data to Achieve Desired Levels of Algorithmic Fairness S28 A Maturity Model for Trustworthy AI Software Development S29 Fairness-aware Configuration of Machine Learning Libraries S30 Diversity-aware fairness testing of machine learning classifiers through hashing-based sampling S31 Exploring and Repairing Gender Fairness Violations in Word Embedding-based Sentiment Analysis Model through Adversarial Patches S32 Efficient white-box fairness testing through gradient search S33 Coverage-Guided Fairness Testing S34 Individual Fairness for Graph Neural Networks: A Ranking based Approach S35 MAFT: Efficient Model-Agnostic Fairness Testing for Deep Neural Networks via Zero-Order Gradient Search S36 A Comprehensive Empirical Study of Bias Mitigation Methods for Machine Learning Classifiers S37 Supporting Many-Objective Software Requirements Decision: An Exploratory Study on the Next Release Problem S38 Multi-objective search for gender-fair and semantically correct word embeddings S39 Software doping analysis for human oversight S40 FairBalance: How to Achieve Equalized Odds With Data Pre-processing S41 The landscape and gaps in open source fairness toolkits S42 Measuring Imbalance on Intersectional Protected Attributes and on Target Variable to Forecast Unfair Classifications Table-A I-2 Research Studies Analyzed in the Systematic Literature Mapping (43-95) # Studies Selected for the Systematic Literature Mapping S43 Fair-SSL: Building fair ML Software with less data S44 Fairea: A model behaviour mutation approach to benchmarking bias mitigation methods S45 White-box fairness testing through adversarial sampling S46 Are My Deep Learning Systems Fair? An Empirical Study of Fixed-Seed Training S47 Algorithmic bias and the value sensitive design approach S48 FairnessLab: A Consequence-Sensitive Bias Audit and Mitigation Toolkit S49 Balanced Fair K-Means Clustering S50 Fair near neighbor search via sampling S51 Testing machine learning algorithms for balanced data usage S52 A responsible machine learning workflow with focus on interpretable models, post-hoc explanation, and discrimination testing S53 Bias in human data: A feedback from social sciences S54 Fairness without harm: Decoupled classifiers with preference guarantees S55 AI Extension of SQuaRE Data Quality Model S56 Data collection and quality challenges in deep learning: a data-centric AI perspective S57 Detecting Discrimination Risk in Automated Decision-Making Systems with Balance Measures on Input Data S58 QuoTe: Quality-oriented Testing for Deep Learning Systems S59 An Algorithmic Framework for Bias Bounties S60 Tiny, Always-on, and Fragile: Bias Propagation through Design Choices in On-device Machine Learning Workflows S61 Experimental study on generating multi-modal explanations of black-box classifiers in terms of gray-box classifiers S62 A Multidimensional Analysis of Social Biases in Vision Transformers S63 De-biasing "bias" measurement S64 Evaluating impact of race in facial recognition across machine learning and deep learning algorithms S65 Evaluating impact of race in facial recognition across machine learning and deep learning algorithms S66 Integrating Fairness in the Software Design Process: An Interview Study With HCI and ML Experts S67 A qualitative exploration of perceptions of algorithmic fairness S68 Assessing the Fairness of AI Systems: AI Practitioners’ Processes, Challenges, and Needs for Support S69 Causal Intervention for Fairness in Multibehavior Recommendation S70 A step toward building a unified framework for managing AI bias S71 Software documentation is not enough! Requirements for the documentation of AI S72 RELIANT: Fair Knowledge Distillation for Graph Neural Networks S73 Certified Machine-Learning Models S74 Abstract Interpretation-Based Feature Importance for Support Vector Machines S75 Slice Tuner: A Selective Data Acquisition Framework for Accurate and Fair Machine Learning Models S76 Proportionally fair clustering S77 The what-if tool: Interactive probing of machine learning models S78 Pros and cons of GAN evaluation measures: New developments S79 Non-functional Requirements for Machine Learning: Understanding Current Use and Challenges in Industry S80 CO*IR: A Greedy and Individually Fair Re-ranker S81 Towards model-based bias mitigation in machine learning S82 Imbalanced data as risk factor of discriminating automated decisions: A measurement-based approach S83 MLCHECK-property-driven testing of machine learning classifiers S84 Multi-objective swarm intelligence approach for bias mitigation in decision-making software S85 Experience: Bridging Data Measurement and Ethical Challenges with Extended Data Briefs S86 FairPlay: A Collaborative Approach to Mitigate Bias in Datasets for Improved AI Fairness S87 How fair are we? From conceptualization to automated assessment of fairness definitions S88 Method for analysis and formation of representative text datasets S89 Diversity Drives Fairness: Ensemble of Higher Order Mutants for Intersectional Fairness of Machine Learning Software S90 Fairness Testing Through Extreme Value Theory S91 Fairness Mediator: Neutralize Stereotype Associations to Mitigate Bias in Large Language Models S92 Building Bridges, Not Walls: Fairness-Aware and Accurate Recommendation of Code Reviewers via LLm-Based Agents Collaboration S93 FairPreprocessor: Better Fairness via Addressing Imbalanced Data through Synthetic Data Generation and Mitigating Biased Labels S94 Beyond the Seeds: Fairness Testing via Counterfactual Analysis of Non-Seed Instances S95 RSUTT: Robust Search Using T-Way Testing