Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Journal of System and Technology
M. Darma Alfero
Politeknik Negeri Padang
Computer Science & IT Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
valuasi Kinerja Lima Algoritma Pembelajaran Mesin dengan SMOTE untuk Klasifikasi Sentimen Komentar YouTube terhadap Kebijakan Purbaya M. Darma Alfero
Journal of System & Technology (SYSTEC) Vol. 2 No. 1 (2026): Journal of System & Technology (June Edition)
Publisher : Jurusan Teknik Elektro, Fakultas Teknik, Universitas Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The evolution of public policy communication in the digital age has encouraged social media to become a deliberative space that openly reflects the dynamics of public perception and sentiment. This paper aims to conduct a comparative analysis of the performance of several machine learning algorithms in classifying the sentiment of YouTube user comments as a representation of public response to policy communication delivered by Minister Purbaya. The research was conducted using web scraping techniques with the YouTube Data API v3 on three public policy videos, with a total of 6,466 comments. The text data were processed using preprocessing, lexicon-based sentiment labeling, TF-IDF feature weighting, and class imbalance handling via the Synthetic Minority Over-sampling Technique (SMOTE). Five machine learning algorithms were applied, comprising SVM, KNN, RF, NB, and LR. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. The results showed that SVM produced the most optimal performance with an accuracy value of 0.78 and a consistent F1-score balance across all sentiment classes. These findings highlight the importance of algorithm suitability, feature representation, and data balancing strategies in developing reliable sentiment analysis models for evaluating data-driven public policy communication.
Co-Authors