: Systemic Lupus Erythematosus (SLE) and Antiphospholipid Syndrome (APS) are two common autoimmune disorders for which the choice of drug regimen is clinically crucial. However, due to drug-drug interactions and individual differences, the therapeutic process faces greater risks. : In this study, we propose a drug recommendation model that combines drug combination frequency, risk assessment, and genetic interaction information with the aim of providing personalized, low-risk treatment options for patients with lupus erythematosus and antiphospholipid syndrome. We extracted drug combination frequencies and drug-gene interaction information from data sources, such as the MIMIC-III clinical database, Drug Bank, and Gene Expression Omnibus. The model comprehensively evaluates the frequency of drug combinations, the risk level, and the gene interaction information through a greedy algorithm to recommend the optimal drug alternatives. : The experimental results show that the model is able to effectively reduce the potential risk between drugs while ensuring the drug treatment effect. In addition, the performance evaluation of the drug recommendation model shows that the model performs well under different drug combinations and clinical scenarios, and can provide clinicians with effective drug substitution suggestions. : This study provides an important theoretical basis and technical support for advancing the realization of personalized therapy and precision medicine.