Published July 30, 2019 | Version v1

Forecast Model for Student

Description

Class and prediction of Student's performance in exam are the standard demanding situations for educators. Diverse conventional records mining strategies consisting of selection tree and association guidelines had been used to carry out class. In latest years, the fast development of synthetic intelligence and deep learning set of rules furnished another method for wise classification and end result prediction. On this paper, a studies on how to use tensorflow artificial intelligence engine for classifying students' performance and forecasting their destiny universities degree application is studied. An appropriate and correct forecast is essential for imparting spark off recommendation to student on software and university choice. For a more complete attention of an all rounded elements, the deep gaining knowledge of model analysed no longer most effective the conventional academic performance such as mathematic, Chinese language, english, physics, chemistry, biology and history, but additionally non-educational overall performance together with provider, behaviour, recreation and artwork. A few parameters in tensorflow engine together with the range of intermediate nodes and variety of deep gaining knowledge of layers are adjusted and as compared. With a information set of hundreds students, 75% of those information are used as the schooling information and 25% are used as the checking out statistics, the accuracy ranged from eighty% to 91%. The surest configuration of the tensorflow deep mastering model that achieves highest prediction accuracy is determined. This look at decided the elements affecting the accuracy of the prediction version.

Files

5955.pdf

Files (305.2 kB)

Name Size
md5:5c15f8ebe3c2f8437e4dcea289355b11
305.2 kB Preview Download

Additional details

Related works

Is version of
Journal article: https://ijsrset.com/IJSRSET196401 (URL)