In this experiment, the National Cancer Institute SEER database was used and 13922 sample data were selected. Based on the basic demographic, clinicopathological and treatment modalities of OCSS patients, survival curves were plotted using the Kaplan-Meier method, the Cox survival analysis model was used to quantitatively assess the risk of death in patients with oral cancer, and the model was evaluated by the consistency index, and the results were presented in columns and lines. The results were presented in a line graph to provide a basis for early judgment and screening of oral cancer patients with poor prognosis. The results showed that age, race, gender, tumor site, radiotherapy, chemotherapy, T-stage, N-stage and M-stage were independent risk factors affecting oral cancer. Clinically, it is recommended to check the status of the cancer cells regularly, focusing on T, N and M stages, and to assess the patient's prognosis.