AUTHOR=Zhang Fang , Zhu Shiben , Chen Siyu , Hao Ziyu , Fang Yuan , Zou Huachun , Cai Yong , Cao Bolin , Zhang Kechun , Cao He , Chen Yaqi , Hu Tian , Wang Zixin TITLE=Application of machine learning for risky sexual behavior interventions among factory workers in China JOURNAL=Frontiers in Public Health VOLUME=Volume 11 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1092018 DOI=10.3389/fpubh.2023.1092018 ISSN=2296-2565 ABSTRACT=Targeted education is key to controlling HIV/STI infection. In the past six months, we built multiple machine learning models to assess the risk of risky sexual behaviors to provide awareness-raising direction. Our study used questionnaire data from 2023 workers at 16 factories in Longhua District CDC, Shenzhen from October 2019 to November 2019. About a quarter of the factory workers had engaged in risky sexual behavior in the past six months. Most of them were Han Chinese (84.53%), hukou in foreign provinces (85.12%) or rural areas (83.19%), with junior high school education (55.37%) and personal monthly income between RMB3,000 (US$417.54) and RMB4,999 (US$695.76) (64.71%), and were workers (80.67%). The random forest model (RF) outperformed all other models in assessing risky sexual behavior in the past six months and provided acceptable performance (accuracy 78%; sensitivity 11%; specificity 98%; PPV 63%; ROC 84%). Machine learning methods have helped assess risky sexual behavior in the past six months and discover crucial risk factors such as leaving home, workload, and exercise. Our assessment models can be incorporated into government or public health departments to provide direction for sexual health promotion or follow-up services.