AI-based author entity disambiguation for promoting fair evaluation of women in science

Allison Dobbins

Pronouns: she/her

Research Mentor(s): Jinseok Kim
Research Mentor School/College/Department: Institute for Social Research / Information
Program: UROPF
Session: Session 3 (11:00am – 11:50am)
Authors: Allison Dobbins, Jinseok Kim

Abstract

Currently, female scholars are considered less productive than their male counterparts due to sampling and entity issues. Previous research on the “Gender Productivity Gap” has used small-sized, biased samples specific to a selected domain. They have also been based on flawed data that doesn’t properly identify female scholars by ignoring name changes after marriage. The purpose of this project is to correct the sampling and identity issues in order to accurately measure female scholar productivity. This will be accomplished by creating large-scale data that trains machine learning algorithms to properly distinguish the entities of female scholars who happen to use different names they have been published under. The goal is to gain meaningful results regarding the “Gender Productivity Gap”. Since this project is still in its early stages, conclusions have not been formed. However, the results might suggest a higher female scholar productivity than previously recorded. These ideas relate to the larger context of how women are regarded in academics as well as society as a whole. Society is based on patriarchy and “evidence” that female scholars are less productive than male scholars only perpetuates this notion in academics. If this project is able to level the playing field, female scholars will be taken more seriously in academics and maybe even society as a whole.

Engineering