Skripsi

KLASIFIKASI KOMENTAR JUDI ONLINE MENGGUNAKAN INDOBERT

---

The rapid growth of social media has enabled the misuse of comment sections as a medium for promoting online gambling. The high volume of comments makes manual moderation ineffective, thus requiring an automated classification system. This study aims to develop a classification system for online gambling comments using the IndoBERT model and to evaluate its performance based on accuracy, precision, recall, and F1-score metrics. IndoBERT was selected due to its ability to understand contextual meaning bidirectionally and its pre-training on a large Indonesian corpus, making it effective in handling informal language variations in social media comments. The dataset used in this study consists of 20,807 Indonesian-language comments and was expanded to 50,000 data points through data augmentation techniques. The dataset was divided into 80% training data, 10% validation data, and 10% testing data. Furthermore, a fine-tuning process was conducted using six hyperparameter scenarios, including learning rate, batch size, and number of epochs. The best performance was achieved in Scenario 5 with a learning rate of 3e-5, batch size of 32, and 3 epochs, resulting in accuracy, precision, recall, and F1-score values of 98.68%.

---

Availability

Detail Information

Series Title

-

Call Number

T1955942026

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xxiii, VI-2 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.310 7

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Pembelajaran Mesin (Machine Learning)

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment