Skripsi

PENERAPAN INDOBERT EMBEDDING DAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) UNTUK KLASIFIKASI UJARAN KEBENCIAN

---

Hate speech on social media is a serious issue that can trigger discrimination and social conflict, highlighting the need for an automated classification system to identify different types of hate speech. This study aims to develop an Indonesian hate speech classification system by combining IndoBERT embeddings as text representations and Support Vector Machine (SVM) as the classification algorithm. The dataset consists of 1,000 Indonesian-language texts categorized into four classes: religious hate speech, physical hate speech, racial hate speech, and gender based hate speech. The experiments were conducted using Linear and Radial Basis Function (RBF) kernels with various values of C and gamma. The best performance was achieved using the RBF kernel with C = 15 and gamma = 0.002, resulting in an accuracy of 84.00%, precision of 81.30%, recall of 77.37%, and an F1-score of 78.07%. The model performs well in classifying explicit hate speech but still faces limitations in handling ambiguous speech.

---

Availability

Detail Information

Series Title

-

Call Number

T1923222026

Publisher

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

Collation

xvi, 138 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

006.312 07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Analisis Data--Support Vector Machine

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment