An Open Dataset of Chinese Name-to-Gender Associations for Gender Prediction in Broad Scientific Research

Identifying an individual's gender based on names is crucial for many scientific studies on gender issues, but it is often complicated by diverse naming conventions globally. Predicting gender from Chinese names poses particular challenges due to unique naming conventions and limited representation in existing datasets. In this study, we introduce a novel dataset comprising 1,051,891 Chinese names in Chinese characters and 96,797 corresponding names in Pinyin from over thirty million Chinese individuals. This dataset includes the frequency of each name's usage by men and women. We validate our dataset using two additional datasets for predicting gender and find that it offers broader name coverage and higher predictive precision compared to existing methods. Overall, this dataset serves as an essential resource for advancing research in China's gender-related studies.