Riyanto Riyanto
Department Information Systems, Universitas Amikom Purwokerto, Indonesia

Published : 1 Documents Claim Missing Document
Claim Missing Document

Found 1 Documents

Application of the Vector Machine Support Method in Twitter Social Media Sentiment Analysis Regarding the Covid-19 Vaccine Issue in Indonesia Riyanto Riyanto; Abdul Azis
Journal of Applied Data Sciences Vol 2, No 3: SEPTEMBER 2021
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v2i3.40


According to the Indonesian government, Indonesia has been afflicted by Covid-19 since March 2, 2020. Numerous countries, including Indonesia, have made efforts, but with the spread of perceptions, rumors, and a flood of information into the society regarding vaccines, there are both advantages and disadvantages to vaccines. government-led immunization campaigns. As a result, it is vital to examine public sentiment toward the government's immunization programs. The goal of this study is to ascertain the emotion toward the Covid-19 vaccination in Indonesia based on the classification results. The Support Vector Machine classification technique was employed in this investigation (SVM). The SVM classification method was chosen because it possesses the ability to generalize when it comes to identifying a pattern, excluding the data used in the method's learning phase. Classification with an SVM linear kernel and TF-IDF weighting, as well as data sharing via K-fold cross validation with a value of k=10. Positive and negative classifications are made. Following preprocessing and classification, we determined the f1 values, accuracy, precision, and recall to use as reference values when evaluating the classification. SVM performed well in classifying the data in this investigation, with  f1 = 88.7%, accuracy = 84.4%, precision = 86.2%, and recall = 97%. This value is acceptable, and hence SVM is suitable for identifying sentiment data about the Covid-19 vaccine in Indonesia. Additional study can be conducted with richer tweet data, more thorough preprocessing, and comparison to other classification algorithms to obtain a higher categorization evaluation score.