Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Applied Information Technology and Computer Science (AICOMS)
Adikara Alif Nurrahman
Unknown Affiliation
Computer Science & IT

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

Found 1 Documents
Search

 1 
Opini Publik terhadap Isu Pengoplosan Pertamax di Youtube Menggunakan Metode Naive Bayes Adikara Alif Nurrahman; Moza , Earlando; Md, Ramanda; Rizvi Roshan , Muhamad; Rizky , Ahmad; Irsyad , Hafiz
Applied Information Technology and Computer Science (AICOMS) Vol 4 No 2 (2025)
Publisher : Pengelola Jurnal Politeknik Negeri Ketapang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58466/aicoms.v4i2.1990

Abstract

This study aims to explore public perceptions regarding the issue of Pertamax fuel adulteration, a topic that has sparked widespread discussion on YouTube, by employing sentiment analysis techniques based on the Naive Bayes algorithm. This issue has attracted significant public attention and become a trending topic on social media, particularly on the YouTube platform. The data analyzed in this research consist of user comments responding to the issue. The Naive Bayes algorithm is used to classify sentiments in the comments into three categories: positive, negative, and neutral. To address the imbalanced distribution of data, the Synthetic Minority Over-sampling Technique (SMOTE) is applied. The results show that before applying SMOTE, the model achieved an accuracy of only 48%, with a precision of 0.48, recall of 0.36, and an F1-score of 0.41 for the negative category, as well as a precision of 0.48, recall of 0.56, and an F1-score of 0.52 for the positive category. After implementing SMOTE, the model's accuracy increased significantly to 88%, with a precision of 0.91, recall of 0.93, and an F1-score of 0.92 for the negative category. For the positive category, precision improved to 0.80, although recall decreased to 0.75, yielding an F1-score of 0.77. The average precision, recall, and F1-score (macro average) after applying SMOTE reached 0.85, 0.84, and 0.85, respectively, representing a substantial improvement compared to the results before SMOTE. This study highlights the importance of using SMOTE to enhance sentiment analysis accuracy, particularly in addressing class imbalance issues within the dataset.
Co-Authors Irsyad , Hafiz Md, Ramanda Moza , Earlando Rizky , Ahmad Rizvi Roshan , Muhamad