Authors:
I. Wayan Suasnawa
;
I. Gusti Ngurah Bagus Caturbawa
;
I. Gede Suputra Widharma
;
Anak Agung Ngurah Gde Sapteka
;
I. Indrayana
and
I. Sunaya
Affiliation:
Department of Electrical Engineering, Politeknik Negeri Bali, Badung, Bali, Indonesia
Keyword(s):
Online Learning, Machine Learning, Naïve Bayes, Support Vector Machine.
Abstract:
The Covid-19 pandemic situation presents a new phenomenon in the world of education. In this situation, it is not possible to conduct face-to-face learning so that online learning becomes the main choice. The online learning method certainly has advantages and disadvantages. There are many comments, both pros and cons regarding the implementation of this online learning. People’s sentiments can be grouped into three, those who feel that the implementation of online learning is able to provide a good solution (positive), those who consider it not an effective solution (negative), and those that are not both (neutral). In this study, the data used in the social media Twitter. In this study, the Naive Bayes classifier and the Support Vector Machine will be used to obtain sentiment analysis on the implementation of online learning during the pandemic. The results of this study indicate that public sentiment is classified into three classes positive, negative and neutral with a precision
level of 0.76 (positive), 0.79 (negative) and 0.92 (neutral) in machine learning using the Naïve Bayes classifier and 0.78 (positive), 0.50 (negative). ) and 0.54 (neutral) on machine learning using the Support Vector Machine classifier. Meanwhile, the accuracy value is above 0.8 for the Naïve Bayes classifier and 0.61 for the Support Vector Machine classifier. The results obtained in machine learning with 2 different classifiers show that the Naïve Bayes classifier has better precision and accuracy than the Support Vector Machine.
(More)