Perbandingan Algoritma Extreme Learning Machine dan Multilayer Perceptron Dalam Prediksi Mahasiswa Drop Out
Abstract
Determined by the university concerned. The high number of drop out students at tertiary institutions can be minimized by policies from tertiary institutions to direct and prevent students from dropping out that detecting at-risk students in the early stages of education is very important to do to keep students from dropping out.
The purpose of this study is to classify and compare the Extreme Learning Machine and Multilater Perceptron algorithms in predicting student drop out. This study uses two algorithms, namely Extreme Learning Machine and Multilater Perceptron which are feedforward artificial neural network learning methods.
The data used is 110 data according to the number of students from class 2012 to 2018. The data is taken from the Doctor of Education Management academic information system. In this case how to predict student drop out using the variables Gender, Working Status, Family Status, Age, Semester 3 GPA, Comprehensive Examination, Dissertation Progress, and Publications. The results of the Extreme Learning Machine classification based on a ratio of 80:20 get an accuracy of 95% with a hidden layer of 20 and a Mean Squared Error value of 0.369. Whereas the Multilater Perceptron with the same ratio gets 91% accuracy. From the two models used, it shows that the two artificial neural network algorithms can produce good performance in predicting drop out students.
References
cccc
Copyright (c) 2023 Muhammad Ibnu Saad Saad
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).