Gendered Algorithms

Image for Gendered Algorithms

More Information

mcds@unimelb.edu.au

Event recording available soon

Do hiring algorithms discriminate against CVs with feminine language?

Project

In the past decade, the adoption of predictive systems in recruitment continues to rise. Central to this are the presumptions that predictive hiring systems enable not only high efficiency but also impartiality: unbiased calculations that lead to mathematical predictions. However, if predictive hiring systems make decisions based on historical data, and data reflects society and its historical biases and assumptions, then these systems are not immune to bias. On the contrary, they have been shown to perpetuate, and sometimes exacerbate societal biases.

Our project, "Gendered Algorithms", builds upon findings from socio-psychological research that corporate recruitment often discriminates against women. With the rising adoption of automation in corporate decisions, we complement extant research by shedding light on how gendered language may exacerbate biased recruitment decisions.

In this talk

This talk discussed our preliminary findings and work in progress from our MCDS Seed Funded Project, "Gendered algorithms".

Briefly, we covered how and to what extent predictive hiring systems can discriminate against women and CVs with feminine language, via an NLP-centric analysis of authentic real-world CVs, specifically:

1.the impetus for our current work, with an introduction from our earlier research projects.

2.contemporary considerations for data collection, including the state of publicly-available CV datasets and desiderata for crowdsourcing.

3.our computational social science-guided investigation on crowdsourced CV data (forthcoming in the Fifth Workshop on NLP and Computational Social Science (NLP+CSS), EMNLP 2022).

This was a hybrid event.

Researchers

Prof Leah Ruppanner

Leah Ruppanner is a Professor of Sociology and Founding Director of The Future of Work Lab at the University of Melbourne. She was previously a Director of The Policy Lab at the University of Melbourne. Her research investigates gender and its intersection to inequalities, technologies and policies. More about Prof Ruppanner.

Dr Marc Cheong

Marc is a Senior Lecturer - Information Systems (Digital Ethics); Senior Fellow (at Melbourne Law School); and an Honorary Burnet Institute Senior Fellow. He is interested in the intersection of technology (big data, social media, etc) and philosophy (existentialism, ethics, epistemology, and Experimental Philosophy). He has a strong background in data science and social media analysis, and he is one of the early pioneers in Twitter research, having completed his PhD on 'Inferring Social Behavior and Interaction on Twitter by Combining Metadata about Users & Messages'. His current research deals with the philosophy (ethics, epistemology, and phenomenology) behind contemporary social media usage and social networking trends. More about Dr Cheong.

Dr Lea Frermann

Lea Frermann is a Lecturer in the School of Computing and Information Systems, in the field of natural language processing (NLP). Her research interests focus on improving automatic understanding of long and complex texts (books, movie scripts), with help of access to common sense knowledge. She is also interested in using machine learning for a deeper understanding of human language processing, and in using such insights to improve automatic language understanding. More about Dr Frermann.

Sheilla Njoto

Sheilla is currently a Research Project Manager at The Policy Lab, leading a team of researchers to conduct research on how artificial intelligence algorithms reconceptualise the meaning of fairness and discrimination, especially in relation to gender. More about Ms (soon to be Dr) Njoto.