Analisis Sentimen Masyarakat Mengenai Kenaikan Harga BBM Pada Komentar Youtube Dengan Metode Gaussian Naïve Bayes-Submit Journal/Ko
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Kalimantan
Author
Prasetio, Bagus
Subject
T Technology (General)
Datestamp
2022-07-21 23:54:24
Abstract :
Youtube is the largest video platform in the world with a total of 1.5 billion users in 2018. Based on the increase in Youtube users in Indonesia and freedom of opinion, this can be approached to find out the opinions of Youtube users on a conflict or major event that is currently taking place. One of them is regarding the increase in fuel prices which is one of the impacts of the tensions between Russia and Ukraine which made world crude oil prices skyrocket to above US$ 100 per barrel. Based on these problems, the authors conducted research related to sentiment analysis from Youtube user comments regarding the increase in fuel prices. This study uses the Gaussian Naïve Bayes method on sentiment analysis regarding the increase in fuel prices through Youtube user comments with the extraction of the FastText feature to see the effect on the accuracy results obtained. The experiment was carried out using 3053 datasets with the distribution of training data and 8:2 test data using a word embedding FastText and Bag of Word as comparisons. The experiment was carried out with a combination of different vector dimensions in the FastText language model with dimensions of 100, 200, and 300 as parameters used in the process of making the language model. Based on the results of the research that has been carried out, the highest accuracy value was obtained in the experiment using a dataset without stopword filtering and language model fasttext size 100 with 74% accuracy, 64% precision, 54% recall, and 58% f1-score. The number of sentiments labeled negative indicates that the response or public opinion is more inclined towards rejection of the government's policy of increasing fuel prices.
Youtube is the largest video platform in the world with a total of 1.5 billion users in 2018. Based on the increase in Youtube users in Indonesia and freedom of opinion, this can be approached to find out the opinions of Youtube users on a conflict or major event that is currently taking place. One of them is regarding the increase in fuel prices which is one of the impacts of the tensions between Russia and Ukraine which made world crude oil prices skyrocket to above US$ 100 per barrel. Based on these problems, the authors conducted research related to sentiment analysis from Youtube user comments regarding the increase in fuel prices. This study uses the Gaussian Naïve Bayes method on sentiment analysis regarding the increase in fuel prices through Youtube user comments with the extraction of the FastText feature to see the effect on the accuracy results obtained. The experiment was carried out using 3053 datasets with the distribution of training data and 8:2 test data using a word embedding FastText and Bag of Word as comparisons. The experiment was carried out with a combination of different vector dimensions in the FastText language model with dimensions of 100, 200, and 300 as parameters used in the process of making the language model. Based on the results of the research that has been carried out, the highest accuracy value was obtained in the experiment using a dataset without stopword filtering and language model fasttext size 100 with 74% accuracy, 64% precision, 54% recall, and 58% f1-score. The number of sentiments labeled negative indicates that the response or public opinion is more inclined towards rejection of the government's policy of increasing fuel prices.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
