Distance learning has emerged as a response to the Covid-19 pandemic, providing students with a new approach to learning. Online learning platforms, utilizing information technology, have become essential in connecting students and teachers. Online tutoring applications offer valuable supplementary educational materials, with various features to support the learning process. Analyzing sentiment on Twitter regarding these online tutoring applications is crucial in determining the best options for students. This study aims to develop an analytical model using the Support Vector Machine (SVM) for online tutoring applications during the Covid-19 pandemic. The research focuses on analyzing positive and negative sentiments within Twitter data, utilizing the Support Vector Machine (SVM) method. The training phase involved 800 manually labeled tweets, consisting of 400 positive and 400 negative sentiments. For testing, 23,511 labeled data points were used. The training data achieved an accuracy of 91.81%. The research successfully achieved an accuracy rate of 90.62% for training and 91% for testing.
Copyrights © 2022