<p>Brain Lower Grade Glioma is a type of brain tumor that shows considerable variation in patient survival and clinical outcomes. Current prognostic studies often rely on individual modeling techniques or limited data integration, which can reduce interpretability and clinical relevance. This study aims to develop an integrated and interpretable framework for identifying key prognostic factors and estimating survival outcomes. A multi-step analytical framework was employed by evaluating multiple machine learning classifiers to predict overall survival status, among which the Support Vector Machine achieved the highest accuracy. Explainable analysis was used to identify the most influential genes, whose biological relevance was validated using protein interaction networks, gene ontology and pathway analysis. Parametric survival time models based on different distributions were applied, and the optimal model was selected using information criteria to estimate survival duration and assess predictor effects. The analysis identified genes such as MKI67, LATS2 and PRDM16, along with diagnosis age, as significant determinants of survival. The selected survival model enabled effective stratification of patients into high-risk and low-risk groups with clearly distinct survival patterns. The novelty of this study lies in the development of a unified and interpretable survival prediction framework that combines machine learning based classification, explainable gene identification, biological validation and parametric survival time modeling within a single pipeline. This approach enhances interpretability and provides clinically meaningful risk stratification, supporting precision oncology applications.</p>