Skripsi

KLASIFIKASI KOMENTAR JUDI ONLINE PADA TIKTOK MENGGUNAKAN METODE LONG SHORT-TERM MEMORY

---

TikTok social media has evolved into one of the most popular digital platforms; however, its comment sections are frequently misused for the covert promotion of online gambling. The text disguise patterns employed by perpetrators render manual moderation mechanisms difficult and inefficient. This study performs text classification on online gambling comments. The method employed is Long Short-Term Memory (LSTM) as the model, selected for its superiority in handling long-term dependencies in text data, along with Word2Vec as the word embedding. The dataset consists of 19,602 entries, divided into 72% training data, 8% validation data, and 20% testing data. After conducting 12 hyperparameter configuration experiments consisting of learning rate, batch size, and the number of epochs using the brute force method, the best configuration result was obtained in Scenario 6. This scenario utilized a learning rate of 0.001, a batch size of 64, and 25 epochs, yielding an accuracy of 92.73%, precision of 92.80%, recall of 90.66%, and an F1-score of 92.73%.

---

Availability

Detail Information

Series Title

-

Call Number

T1897252025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xvi, 162 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

364.172 07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Judi Online--Tiktok

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment