Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Denny Manuel Yeremia Sinurat
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

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

Found 1 Documents
Search

 1 
Analisis Sentimen Terhadap Kenaikan Cukai Rokok pada Media Sosial Twitter menggunakan Algoritma Naive Bayes Classifier Denny Manuel Yeremia Sinurat; Dian Eka Ratnawati; Dwija Wisnu Brata
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 7 No 1 (2023): Januari 2023
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Social media is a place for people to convey opinions, and also criticism. The social media that is often used by Indonesian people to express their opinions is Twitter. Twitter can be a forum for Indonesian people to express their opinions and aspirations on certain matters, such as policies made by the government. One of the policies that has sparked public discussion on social media Twitter is increasing cigarettes tax by an average of 10% starting in 2023. This policy has drawn pros and cons from the public, therefore an analysis of public sentiment regarding the cigarette tax increase policy can be carried out. This study will analyze public sentiment contained in the tweets of Twitter users about the increase cigarette tax into the classification of positive and negative sentiments. The execution process is implemented in Google Colaboratory with the Naive Bayes Classifier algorithm. Processes in the sentiment analysis which include data collection, manual labeling, text preprocessing, TF-IDF weighting, data balancing with the Synthetic Minority Oversampling Technique (SMOTE), data validation using k-fold cross validation, and test the classification results with the confusion matrix. The best accuracy results of 74% were obtained by using the data balance SMOTE results with a comparison of training data: test data of 90%:10%, text preprocessing, TF-IDF weighting, and use Multinomial Naive Bayes. The highest cross-validation score obtained was 78% with an average of 73%. Based on these results, the Naive Bayes Classifier is quite good as an alternative for sentiment analysis.
Co-Authors Dian Eka Ratnawati Dwija Wisnu Brata