Grant: Discovery Launch Supplement
University of Calgary
$12,500.00
Breaking Bias: Fostering Fairness Testing in Software Engineering
EF 2024-25
Résumé
Background. Currently, software holds significant influence across numerous facets of modern society, spanning work, education, politics, leisure, and beyond. However, software engineering does not fully reflect the diversity of the society it serves, leading to unintentional discrimination against certain groups (e.g., Black people, LGBTQIA+ individuals, or Women), including biased results in internet searches, exclusion from hiring systems, disadvantage in financial and health services and more. In this context, prioritizing fairness in software development is essential for mitigating these issues and guaranteeing that individuals in our society have an equitable experience devoid of discrimination and bias when engaging with technological solutions. Objective. This research program aims to explore fairness in software engineering and create a comprehensive framework for fairness testing. The framework will assist software professionals in comprehending fairness requirements, designing and conducting fairness tests, managing fairness issues, and mitigating algorithmic bias, as it will include: a test oracle that will serve as a guide for test engineers to validate the behaviors of data-driven systems considering sensitive fairness attributes (e.g., gender, ethnicity, sexual orientation); a method for evaluating fairness test adequacy, emphasizing coverage and metrics; a testing tool to assist practitioners in dealing with fairness-related flaws; and a strategy to effectively incorporate testing fairness within software projects. Method. Given the nature of the problem under investigation, this research program calls for the implementation of a mixed-method strategy. A constructivist grounded theory will be one of the main strategies applied in this research. This method will be used to combine the background available in the scientific literature on algorithmic bias and the data collected from the field (e.g., software companies, software teams, and software professionals) to build on the existing data about the problem generate new useful knowledge, thus, mitigating the risks of simply repeating findings from prior work. Following this, design science - a problem-solving empirical approach that emphasizes the development and evaluation of artifacts - will be used to build the framework (including models, guidelines, and tools) and introduce it into the software industry. Then, experiments will be implemented to validate the applicability of the research artifacts within the software industry. Outcomes and Impact. Our research program aims to create a framework comprising a well-defined set of principles, practices, guidelines, and tools, tailored to establish fairness testing as a core aspect of software development. This work has the potential to establish fairness as a standard practice in the software industry, contributing to the reduction of algorithmic biases and diminishing the harm presently inflicted by software in our society.
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
DGECR-2024-00043
Name nom
de Souza Santos, Ronnie
Department département
Schulich School of Engineering
Organization id
11
Institution établissement
University of Calgary
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
2706
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Software engineering
Research subject fr
Génie logiciel
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Breaking Bias: Fostering Fairness Testing in Software Engineering
Keywords
software fairness; software development; fairness testing; software practices; algorithmic discrmination; bias
Application summary
Background. Currently, software holds significant influence across numerous facets of modern society, spanning work, education, politics, leisure, and beyond. However, software engineering does not fully reflect the diversity of the society it serves, leading to unintentional discrimination against certain groups (e.g., Black people, LGBTQIA+ individuals, or Women), including biased results in internet searches, exclusion from hiring systems, disadvantage in financial and health services and more. In this context, prioritizing fairness in software development is essential for mitigating these issues and guaranteeing that individuals in our society have an equitable experience devoid of discrimination and bias when engaging with technological solutions. Objective. This research program aims to explore fairness in software engineering and create a comprehensive framework for fairness testing. The framework will assist software professionals in comprehending fairness requirements, designing and conducting fairness tests, managing fairness issues, and mitigating algorithmic bias, as it will include: a test oracle that will serve as a guide for test engineers to validate the behaviors of data-driven systems considering sensitive fairness attributes (e.g., gender, ethnicity, sexual orientation); a method for evaluating fairness test adequacy, emphasizing coverage and metrics; a testing tool to assist practitioners in dealing with fairness-related flaws; and a strategy to effectively incorporate testing fairness within software projects. Method. Given the nature of the problem under investigation, this research program calls for the implementation of a mixed-method strategy. A constructivist grounded theory will be one of the main strategies applied in this research. This method will be used to combine the background available in the scientific literature on algorithmic bias and the data collected from the field (e.g., software companies, software teams, and software professionals) to build on the existing data about the problem generate new useful knowledge, thus, mitigating the risks of simply repeating findings from prior work. Following this, design science - a problem-solving empirical approach that emphasizes the development and evaluation of artifacts - will be used to build the framework (including models, guidelines, and tools) and introduce it into the software industry. Then, experiments will be implemented to validate the applicability of the research artifacts within the software industry. Outcomes and Impact. Our research program aims to create a framework comprising a well-defined set of principles, practices, guidelines, and tools, tailored to establish fairness testing as a core aspect of software development. This work has the potential to establish fairness as a standard practice in the software industry, contributing to the reduction of algorithmic biases and diminishing the harm presently inflicted by software in our society.