Objective To analyze the current status and influencing factors of social isolation in patients with colorectal cancer after stoma surgery, and to construct a risk prediction model for social isolation in these patients.  Methods A total of 507 stoma patients who visited the Department of General Surgery (Colorectal Surgery), Oncology Department and Wound Stoma Clinic of a tertiary Grade A hospital in Jinzhou City from March 2025 to January 2026 were selected as the research subjects by convenient sampling. The data were randomly divided into a training set and a validation set at a ratio of 7:3. LASSO algorithm and Logistic regression analysis were combined to screen the risk factors. Five machine learning algorithms, namely Logistic Regression (LR), Support Vector Machine (SVM), Naive Bayes, K-Nearest Neighbor (KNN) and Decision Tree, were used to construct the prediction models for social isolation in stoma patients. The area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, precision and F1 score were used to evaluate and compare the predictive performance of the models. Finally, the optimal risk prediction model with the best predictability and practicability was selected for SHapley Additive exPlanations (SHAP) analysis to demonstrate the importance of predictive factors.  Results Among the 507 patients with colorectal cancer after stoma surgery, 163 cases suffered from social isolation, with an incidence rate of 32.3%. Among the five machine learning models, the LR model showed the best performance, with an AUC of 0.813 (95%CI: 0.746–0.882), a sensitivity of 0.544, an accuracy of 0.724, a precision of 0.660, a specificity of 0.832 and an F1 score of 0.596. The SHAP bar chart showed that the top four influencing factors were gender, place of residence, social support and psychological vulnerability.  Conclusion The LR model has the best performance in predicting the risk of social isolation in patients with colorectal cancer after stoma surgery, which can provide a screening tool for clinical workers to conduct early identification of high-risk groups.