Predicting Students' Performance in Final Examination using Deep Neural Network

The academic result is the most important thing in a student's career. This result depends on their academic performance and many other factors. Educational data mining can help both students and institutions develop their academic performance. For analysis of their performance, we can use new techniques Deep Learning, Convolution Neural Networks, Data Clustering, Optimization Algorithms, etc. In machine learning. Using Deep Learning, we will predict the student’s performance yearly in the form of CGPA and compare that with the real CGPA. A real dataset can boost the prediction performance. We used a real dataset from the Institute of Science, Trade & Technology (ISTT). We used a total of 18 data factors to predict the performance and the data factors are: Class Performance, Test Marks, Class Attendance, Due Time Assignment Submission, Lab Performance, Previous Semester Result, Family Education, Freelancer, Relationship with Faculty, Study Hours, Living Area, Social Media Attraction, Extra-Curricular Activity, Drug Addiction, Financial Support from Family, Political Involvement, Affair & Year Final Result.