Innovative ideas and gender inequality

This paper analyzes the recognition of women's innovative ideas. Bibliometric data from research in economics are used to investigate gender biases in citation patterns. Based on deep learning and machine learning techniques, one can (1) establish the similarities between papers (2) build a link between articles by identifying the papers citing, cited and that should be cited. This study finds that, on average, omitted papers are 15%-20% more likely to be female-authored than male-authored. This omission bias is more prevalent when there are only males in the citing paper. Overall, to have the same level of citation as papers written by males, papers written by females need to be 20 percentiles upper in the distribution of the degree of innovativeness of the paper.