GRADMETRIC-STUDENT CGPA PREDICTOR

In higher technical education, academic performance is usually assessed after the completion of a semester, leaving little scope for improvement within the same academic cycle.Such delayed evaluation often results in increased failure rates and limited placement opportunities.This paper presents a predictive framework designed to estimate a student's final Cumulative Grade Point Average (CGPA) using mid-semester academic indicators, including attendance records, internal assessment scores, and laboratory participation.Regressionbased ensemble learning techniques are applied to capture complex performance patterns.In addition to prediction, the system incorporates an intervention mechanism that assigns academically vulnerable students to a structured technical recovery program.Experimental observations indicate that early identification of academic risk can improve final performance and reduce overall academic stagnation.