Objective To analyze the influencing factors of moderate and severe malnutrition in oral cancer patients after radical resection and to build a comprehensive prediction model. Method A total of 195 patients who underwent radical resection of oral cancer treated in Capital Medical University Affiliated Beijing Stomatological Hospital from January 2022 to January 2024 as the research subjects. Two weeks after surgery they were divided into a moderate and severe malnutrition group of 117 cases and a non malnutrition group of 78 cases according to their nutritional status. The risk factors for moderate and severe malnutrition in oral cancer patients after radical resection were analyzed and a regression model and decision tree model were constructed to comprehensively evaluate postoperative moderate and severe malnutrition. Result According to binary Logistic regression analysis clinical stage Ⅲ OR = 6. 114 95%CI = 1. 667-22. 423 clinical stage Ⅳ OR = 6. 537 95%CI = 1. 656-25. 797 oral treatment history OR = 3. 387 95%CI = 1. 388- 8. 263 love of chewing betel nut OR = 3. 783 95% CI 1. 550- 9. 232 tracheotomy OR = 3. 120 95% CI = 1. 196-8. 139 postoperative radiotherapy OR = 21. 625 95%CI 6. 457-72. 421 postoperative depression OR = 3. 832 95% CI = 1. 506 - 9. 751 postoperative sleep disorders OR = 3. 940 95% CI = 1. 600 - 9. 704 and postoperative swallowing disorders OR= 3. 474 95%CI = 1. 427-8. 462 are risk factors for moderate and severe malnutrition in patients with oral cancer undergoing radical resection P<0. 05 . The model's Hosmer Lemeshow fitting test showed a chi square of 5. 613 and a P-value of 0. 691. ROC curve analysis revealed that the AUC of the model for predicting moderate and severe malnutrition in oral cancer patients after radical resection was 0. 904 95%CI = 0. 862-0. 945 with a sensitivity of 81. 2% specificity of 83. 3% Youden index of 0. 645 and an accuracy of 82. 1% in practical applications. A decision tree risk model for postoperative moderate and severe malnutrition in oral cancer patients undergoing radical resection was constructed with the occurrence of postoperative moderate and severe malnutrition as the dependent variable and risk factors as independent variables. The misjudgment rate was 21. 0% and the prediction accuracy was 79. 0%. Conclusion Malnutrition and clinical stage oral treatment history preference for chewing betel nut tracheotomy and postoperative radiotherapy depression sleep disorders and swallowing disorders are related to patients with oral cancer undergoing radical resection. Logistic regression models and decision tree models have high predictive value for moderate and severe malnutrition in patients with oral cancer undergoing radical resection.