Objective: To explore the case mix mode of hospitalization expenses for chemotherapy patients and to construct the big data diagnosis-intervention package (DIP) ssisted catalogue. Methods: The information of 27504 patients undergoing chemotherapy in a hospital from 2019 to 2021 was collected. One-way analysis of variance and stepwise regression analysis were used to screen the influencing factors of hospitalization expenses as classification node variables. Cluster analysis and decision tree model were used to construct DIP assisted catalogue for chemotherapy patients. Results: Tumor types，comorbidities/complications，age，and tumor metastasis were used as the classification nodes affecting the hospitalization cost of chemotherapy patients. 20 DIP disease combinations were constructed. The CV value of most of the case combinations remained around 0.6，and the RIV was 12.51 %，with good intra-group homogeneity and inter-group heterogeneity. Conclusion: Using decision tree model for case combination method to construct DIP assisted catalogue for chemotherapy patients is scientific and reasonable，which can provide references for DIP payment for chemotherapy patients.