Diversity in Software Engineering Conferences and Journals
Authors:
Aditya Shankar Narayanan,
Dheeraj Vagavolu,
Nancy A Day,
Meiyappan Nagappan
Abstract:
Diversity with respect to ethnicity and gender has been studied in open-source and industrial settings for software development. Publication avenues such as academic conferences and journals contribute to the growing technology industry. However, there have been very few diversity-related studies conducted in the context of academia. In this paper, we study the ethnic, gender, and geographical div…
▽ More
Diversity with respect to ethnicity and gender has been studied in open-source and industrial settings for software development. Publication avenues such as academic conferences and journals contribute to the growing technology industry. However, there have been very few diversity-related studies conducted in the context of academia. In this paper, we study the ethnic, gender, and geographical diversity of the authors published in Software Engineering conferences and journals. We provide a systematic quantitative analysis of the diversity of publications and organizing and program committees of three top conferences and two top journals in Software Engineering, which indicates the existence of bias and entry barriers towards authors and committee members belonging to certain ethnicities, gender, and/or geographical locations in Software Engineering conferences and journal publications. For our study, we analyse publication (accepted authors) and committee data (Program and Organizing committee/ Journal Editorial Board) from the conferences ICSE, FSE, and ASE and the journals IEEE TSE and ACM TOSEM from 2010 to 2022. The analysis of the data shows that across participants and committee members, there are some communities that are consistently significantly lower in representation, for example, publications from countries in Africa, South America, and Oceania. However, a correlation study between the diversity of the committees and the participants did not yield any conclusive evidence. Furthermore, there is no conclusive evidence that papers with White authors or male authors were more likely to be cited. Finally, we see an improvement in the ethnic diversity of the authors over the years 2010-2022 but not in gender or geographical diversity.
△ Less
Submitted 24 October, 2023;
originally announced October 2023.
A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness
Authors:
Tiantian Feng,
Rajat Hebbar,
Nicholas Mehlman,
Xuan Shi,
Aditya Kommineni,
and Shrikanth Narayanan
Abstract:
Speech-centric machine learning systems have revolutionized many leading domains ranging from transportation and healthcare to education and defense, profoundly changing how people live, work, and interact with each other. However, recent studies have demonstrated that many speech-centric ML systems may need to be considered more trustworthy for broader deployment. Specifically, concerns over priv…
▽ More
Speech-centric machine learning systems have revolutionized many leading domains ranging from transportation and healthcare to education and defense, profoundly changing how people live, work, and interact with each other. However, recent studies have demonstrated that many speech-centric ML systems may need to be considered more trustworthy for broader deployment. Specifically, concerns over privacy breaches, discriminating performance, and vulnerability to adversarial attacks have all been discovered in ML research fields. In order to address the above challenges and risks, a significant number of efforts have been made to ensure these ML systems are trustworthy, especially private, safe, and fair. In this paper, we conduct the first comprehensive survey on speech-centric trustworthy ML topics related to privacy, safety, and fairness. In addition to serving as a summary report for the research community, we point out several promising future research directions to inspire the researchers who wish to explore further in this area.
△ Less
Submitted 16 April, 2023; v1 submitted 17 December, 2022;
originally announced December 2022.
