GARD: Gender difference analysis and recognition based on machine learning
Shiwen He,
Jian Song,
Yeyu Ou,
Yuanhong Yuan,
Xiaojie Zhang,
Xiaohua Xu
Affiliations
Shiwen He
School of Computer Science and Engineering, Central South University, Changsha 410083, China; National Mobile Communications Research Laboratory, Southeast University, Nanjing 210096, China; The Purple Mountain Laboratories, China; Corresponding author at: School of Computer Science and Engineering, Central South University, Changsha 410083, China.
Jian Song
School of Computer Science and Engineering, Central South University, Changsha 410083, China
Yeyu Ou
School of Computer Science and Engineering, Central South University, Changsha 410083, China
Yuanhong Yuan
Emergency Center of Hunan Children’s Hospital, Changsha 410007, China; Corresponding author.
Xiaojie Zhang
Department of Psychiatry, The Second Xiangya Hospital, Central South University, Changsha 410011, China; National Clinical Research Center for Mental Disorders, Changsha, 410011, China
Xiaohua Xu
Department of Endocrinology, Nanjing First Hospital, Nanjing Medical University, Nanjing 210012, China
In recent years, intelligent diagnosis and intelligent medical treatment based on big data of medical examinations have become the main trend of medical development in the future. In this paper, we propose a method for analyzing the difference between males and females in medical examination items (medical attributes) and find that males and females of different ages have differences in medical attributes. Then, the cluster analysis method is used to further analyze the differences between male and female in medical examination items, such that some common important attributes (CIAs) that can be used for gender recognition are found within a specific age range. Following, we propose two gender recognition models (GRMs) by using the found CIAs to identify the gender. A large number of experimental results are provided to validate the effectiveness of the proposed GRMs. Experimental results show that the medical attributes with a large value of difference really contribute to gender recognition. Within a certain age range, such as 17 to 51 years old, the proposed GRM can reach 92.8% accuracy using only six medical attributes.
