Education institutions are generating huge volumes of academic data at both student-level and academic performance level. Manual analysis of this data to identify students who might have trouble academically is time consuming for educators and also tends to lack efficiency. If identification of those students are delayed, the negative impact arises from those students having difficulty academically could lead to poor academic outcomes for that student population as a whole.

To address these issues, this research introduces a Student Performance Prediction System using Machine Learning that can support educators in predicting academic outcomes for students and providing timely academic interventions.

The proposed model will analyze data collected from multiple sources of academic data including, but not limited to, attendance, study hours, internal assessment scores, assignment outcomes, and previous academic history; and utilize various Machine Learning algorithms as well as Data Analysis Techniques, to identify trends and relationships between these different data sets and their associated academic outcomes for the student population.

Once all the data is analyzed, the Machine Learning Model will predict how well a given student will likely perform on upcoming examinations. Educators may use these predictions to identify individual students that may need extra academic support based on their predicted performance level. This predictive analysis provides educational institutions with the opportunity to provide early academic intervention strategies, improve teaching strategies, and ultimately enhance the academic performance of their student body