In previous studies, we proposed a risk evaluation model for cervical cancer based on several well-known environmental contributors, such as infection with high-risk genotype(s) of human papillomavirus (HPV),3–5 young age at first parturition,6 and low educationlevel ofthe subject andtheir spouse.7,8 Emerging evidence suggests that human leukocyte antigen (HLA) class II alleles are associated with cervical cancer.9 In this study, we combined HLA class II alleles with several risk factors to establish a risk evaluation model using multiple logistic regression analysis and an artificial neural network. This new evaluation model could significantly improve the accuracy of risk evaluation for cervical cancer.