Forecasting of Student’s Performance and Future Development using Deep Learning
Stratification and prognostic of college students’ accomplishment in exam are the unexceptional summons for educators. Assorted mining tactics inclusive of Naïve Bayes, Decision tree, have been used to find the attributes affecting the student’s performance. Mining methods considered only academic performance as parameters. The existing work contemplates to outlook student acquirement in education make avail of deep learning techniques. Real-world data (e.g. school related traits and student grades) was assembled by plying school proclaim and opinion polls. Although student attainment is immensely persuaded by bygone academic summing up, an interpretive examination has divulged that supplementary pertinent traits exists (e.g. parent’s education, ordinal of non-appearance). Deep Learning focus at draw out high-level apprehension from raw data, proffer automated contraptions that can succor the education dominion. With the collected set of data of 2000 college students, seventy-five percent of it is used for training and twenty-five percent is used for data testing, the accuracy ranged around eighty to ninety-one percent. The contour of the deep learning facsimile which gives maximum accuracy for prediction is determined.