Prediction Analysis Student Graduate Using Multilayer Perceptron

DOI

10.2991/assehr.k.200521.011How to use a DOI?

Keywords

Multilayer Perceptron (MLP), data mining, correctly classified instances, Root Mean Squared Error (RMSE)

Abstract

Student graduation data is a data that is important to the College, especially for the Faculty as well as the courses in question. Acquisition of knowledge in a database (a number of large data) commonly referred to as data mining. This research aims to analyze the student’s graduation predictions that can be done on a fourth semester using Multilayer Perceptron (MLP) classifier which available in WEKA software implementations. Then do the testing and performance comparisons of MLP against Naïve Bayes classification, IBk and Tree J48. Cross Validation and Percentage Split are used as the testing procedure in this research. The parameters in the process of testing using correctly classified instances and Root Mean Squared Error (RMSE). On the mode of Cross Validation, MLP has better performance compared to all contender methods with accuracy of J48 81.82% and the value of the smallest RSME i.e. 0.273. On a Percentage Split MLP mode has the same accuracy value with Naïve Bayes i.e. 92.31%, and the value of the RMSE on the MLP of 0.182.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Mariana Windarti
AU  - Putri Taqwa Prasetyaninrum
PY  - 2020
DA  - 2020/05/22
TI  - Prediction Analysis Student Graduate Using Multilayer Perceptron
BT  - Proceedings of the International Conference on Online and Blended Learning 2019 (ICOBL 2019)
PB  - Atlantis Press
SP  - 53
EP  - 57
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.200521.011
DO  - 10.2991/assehr.k.200521.011
ID  - Windarti2020
ER  -