CVDs is one of the main causes of death in the recent times. Hospital readmissions include a major portion of hospital cases recorded. This study proposes developing and approving a data-driven model to predict readmission risk among patients. A comprehensive dataset was analysed, which included patient demographics, clinical characteristics, drugs, treatment information, and genetic disorders. Data preprocessing, feature engineering, model training, and evaluation were all carried out thoroughly. The findings demonstrate the ability of data mining to accurately predict readmission risk in cardiovascular patients. The constructed Random Forest-based model demonstrated promising performance, allowing for targeted treatments. The findings focus on the importance of data-driven approaches for predicting readmission risk and optimizing healthcare budget allocation. Future studies might investigate the incorporation of these models into clinical workflows to enhance patient outcomes and minimize healthcare expenditures.