MACHINE LEARNING MODEL TO PREDICT THE RISK OF STUDENTS NOT GRADUATING ON TIME

PUTRA, DIONYSIUS ABIRAMA (2023) MACHINE LEARNING MODEL TO PREDICT THE RISK OF STUDENTS NOT GRADUATING ON TIME. Other thesis, UNIVERSITAS KHATOLIK SOEGIJAPRANATA.

[img]
Preview
 Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-COVER_a.pdf
Download (341kB) | Preview
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-BAB I_a.pdf
Restricted to Registered users only
Download (116kB)
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-BAB II_a.pdf
Restricted to Registered users only
Download (240kB)
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-BAB III_a.pdf
Restricted to Registered users only
Download (994kB)
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-BAB IV_a.pdf
Restricted to Registered users only
Download (354kB)
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-BAB V_a.pdf
Restricted to Registered users only
Download (228kB)
[img]
Preview
 Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-DAPUS_a.pdf
Download (336kB) | Preview
[img] Text
16.K1.0052-DIONYSIUS ABIRAMA PUTRA-LAMP_a.pdf
Restricted to Registered users only
Download (223kB)

Abstract

The purpose of this study is creating a program to predict the risk of college student not graduating on time, which is graduating after more than four years of study. The approach to solve this problem is by developing machine learning with Random Forest algorithm. The model is trained and tested using dummy dataset of US student academic records within the first semester to fourth semester. The model is also trained with different number of generated tree. To analyze the result, Confusion Matrix is used to compare the accuracy, precision, misclassification, and recall rates of each trained model. The final result shown that Random Forest model with 91 generated tree has the highest accuracy of 67%, to predict the risk of student not graduating on time.

Item Type: Thesis (Other)
Subjects: 000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: Mr Yosua Norman Rumondor
Date Deposited: 08 Dec 2023 01:50
Last Modified: 08 Dec 2023 01:50
URI: http://repository.unika.ac.id/id/eprint/33806

Actions (login required)

View Item View Item
Unika Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.