KLASIFIKASI SENTIMEN KOMENTAR YOUTUBE MALAKA PROJECT MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) (STUDI KASUS: VIDEO "BISNIS GELAP DOKTER & PERUSAHAAN FARMASI")

Laila Rohmatul I’zzah, NIM.: 22106010007 (2026) KLASIFIKASI SENTIMEN KOMENTAR YOUTUBE MALAKA PROJECT MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) (STUDI KASUS: VIDEO "BISNIS GELAP DOKTER & PERUSAHAAN FARMASI"). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (KLASIFIKASI SENTIMEN KOMENTAR YOUTUBE MALAKA PROJECT MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) (STUDI KASUS: VIDEO "BISNIS GELAP DOKTER & PERUSAHAAN FARMASI")) 22106010007_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (6MB) | Preview
	Text (KLASIFIKASI SENTIMEN KOMENTAR YOUTUBE MALAKA PROJECT MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) (STUDI KASUS: VIDEO "BISNIS GELAP DOKTER & PERUSAHAAN FARMASI")) 22106010007_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (20MB) | Request a copy

Abstract

YouTube comments are unstructured textual data that contain diverse public opinions and require a classification algorithm capable of effectively handling highdimensional text data. This study aims to compare the performance of linear, polynomial, and Radial Basis Function (RBF) kernels in the Support Vector Machine (SVM) algorithm for sentiment classification of comments on the Malaka Project YouTube channel. The dataset was collected through scraping using the YouTube Data API v3 from a video entitled “The Dark Business of Doctors and Pharmaceutical Companies,” resulting in 3,428 comments collected in December 2025. The data were classified into three sentiment categories: positive, negative, and neutral. The research procedure consisted of data preprocessing, sentiment labeling using the InSet Lexicon dictionary, data splitting with an 80:20 ratio, feature extraction using TF-IDF, classification using a multiclass SVM algorithm with the One-Against-All (OvR) approach, and model evaluation using accuracy, precision, recall, F1-score, and confusion matrix metrics. The results show that the linear and RBF kernels achieved the same accuracy score of 0.73, while the polynomial kernel obtained a lower accuracy score of 0.60. However, the linear kernel achieved the highest precision and F1-score of 0.77 for the positive sentiment class. Therefore, it can be concluded that the linear kernel performs better in classifying sentiment in Malaka Project YouTube comments compared to the other two kernels.

Share this knowledge with your friends :

Actions (login required)

View Item