Sentiment Analysis Vaksinasi Covid-19 Pada
Twitter Menggunakan Metode Support Vector Machine(SVM)
Total View This Week0
Total View This Week0
Institusion
Politeknik Negeri Jember
Author
Shofi'i, Aan Nur
Subject
458 - Teknik Informatika
Datestamp
2022-08-26 02:34:19
Abstract :
In 2020, the world community is busy about the presence of a deadly virus that is deadly Coronavirus disease 2019 (COVID-19). Seeing the rapid spread of COVID- 19 that will be caused if it is not resolved, one solution to reduce the rate of virus spread is by making vaccines. This study aims to determine the accuracy of the Support Vector Machine Method on the Covid-19 Vaccination Sentiment Analysis on Twitter. This study uses 1000 data that has been processed, classification of data by training data as much as 800 and then used to classify 200 test data. In this study using one of the kernels from SVM, namely the Polynomial Kernel with the best parameter combination value of degree = 2, = 1, and maximum iteration = 300, and Cross Validation testing for evaluation of the best 80% training data, namely kfold = 10 or as much as 10x, and get the highest level of accuracy in the 7-fold test with an accuracy value of 71%. And testing the 20% test data using the confusion matrix got an accuracy value of 69%.
In 2020, the world community is busy about the presence of a deadly virus that is deadly Coronavirus disease 2019 (COVID-19). Seeing the rapid spread of COVID- 19 that will be caused if it is not resolved, one solution to reduce the rate of virus spread is by making vaccines. This study aims to determine the accuracy of the Support Vector Machine Method on the Covid-19 Vaccination Sentiment Analysis on Twitter. This study uses 1000 data that has been processed, classification of data by training data as much as 800 and then used to classify 200 test data. In this study using one of the kernels from SVM, namely the Polynomial Kernel with the best parameter combination value of degree = 2, = 1, and maximum iteration = 300, and Cross Validation testing for evaluation of the best 80% training data, namely kfold = 10 or as much as 10x, and get the highest level of accuracy in the 7-fold test with an accuracy value of 71%. And testing the 20% test data using the confusion matrix got an accuracy value of 69%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
