The most important liver tumor as well as the primary reason for tumor-associated mortality worldwide is hepatocellular carcinoma (HCC). The long-term outcomes of individuals with HCC and the identification of predictive factors will aid in selecting the best course of action for particular patients. In this research, researchers have utilized machine learning algorithms to determine the significance of the features and predict the patients’ perseverance. During preprocessing, Min-Max Scaling, label encoding, and average imputation were applied. To address class inequalities, SMOTE (Synthetic Minority Over-sampling Technique) was used. SHAP (SHapley Additive Explanations) and LIME (Local Interpretable Model-Agnostic Explanations) were used for feature significance and forecast logic. The LightGBM classifier's assessment yielded an F1 Score of 88%, a precision of 86%, and a recall of 90%. With an accuracy of 88%, the method we suggested outperformed all other approaches currently in use. The combination of SHAP and LIME allowed for better comprehension of forecasting features, essential for clinical usage. The contributions provide tools for improving patient care, including comparison of classifiers, SMOTE application, and feature significance recognition. The suggested machine learning architecture could change how patient outcomes are predicted, leading to more individualized HCC therapies.