Skripsi

KLASIFIKASI UANG KULIAH TUNGGAL (UKT) DI UNIVERSITAS SRIWIJAYA MENGGUNAKAN RANDOM FOREST

---

Penilaian

---

Single Tuition Fee (UKT) system in State Universities (PTN) is a response to the diversity of students' economic conditions in covering tuition fees. This research aims to determine the UKT class of students at Sriwijaya University by using the Random Forest method. This method is an ensemble method of decision trees and was chosen because of its effectiveness in capturing patterns in data and handling overfitting in unbalanced data. The target variable is the UKT class consisting of 8 categories, with a variety of testing parameters, namely n_estimator (number of trees), max_depth (tree depth), and the ratio of test data and training data. The results showed that the combination of 35 trees and max_depth 15 produced the best results with accuracy, recall, and precision of 0.89, and F1-score of 0.88, respectively. Although increasing the number of trees in the Random Forest ensemble improves accuracy, it can affect computation time. The test data and training data ratio of 20:80 resulted in an optimal accuracy of 0.89. Confusion matrix analysis shows that the model can classify UKT class labels well, but there is still difficulty in distinguishing labels that have similar features. Keywords: Random Forest, Classification, Single Tuition Fee (UKT)

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2407003962	T149179	T1491792024	Central Library (Reference)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1491792024

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xiv, V-6 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.310 7

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Uang Kuliah Tunggal
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment