Exploring Weibo users’ attitudes toward lesbians and gays in Mainland China: A natural language processing and machine learning approach

Abstract Informed by the ABC model of attitude, this study employed natural language processing and machine learning techniques to explore the status quo, trends, and features of Weibo users’ attitudes toward individuals who are lesbian or gay based on 1,813,218 Weibo retrieved posts that pertained to lesbians and gays from January 1, 2010, to December 31, 2019. Although the general attitudes, feeling/emotional reactions, and levels of support for the rights of lesbians and gays were identified as relatively positive for Weibo users and became even more so over time, the expressed knowledge of lesbians and gays was relatively inaccurate and the four aspects of attitude varied significantly as a function of individual-level variables (such as gender, age, education, sexual orientation, and verification type). The implications of these findings and suggestions that may improve the public acceptance of lesbians and gays are discussed.