ANALISIS SENTIMEN PENGGUNA TWITER TERHADAP METODE PEMBELAJARAN JARAK JAUH MENGGUNAKAN ALGORITMA NAÏVE BAYES
No Thumbnail Available
Date
2022-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Elfitra
Abstract
The Covid-19 pandemic has caused various impacts, one of which is on educational
activities. The Indonesian government decided to implement learning by the Distance
Learning method. This has become a hot topic of discussion on Twitter. The circulated
tweets contain public opinions on the implementation of Distance Learning. This study
aims to classify sentiments to see the public's view of the Distance Learning
implementation. Data used in this study is obtained by using the keywords of online
lecture, networked lecture, online school, and networked school from October to
November 2020. The sentiment is classified into two classes: positive and negative. The
classification process is carried out by the pre-processing data and word weighting using
the TF/IDF technique. Data classification is done by using the Naïve Bayes Multinomial
algorithm and evaluation of the algorithm is done by using the confusion matrix method.
The final result of the classification process yields a good accuracy rate of 83%, precision
of 33.33% and recall of 6.25%. The results show that netizens do not fully agree with the
implementation of Distance Learning. It is showed by negative sentiments are 97% and
positive sentiments are 3%.
Description
Keywords
Distance Learning, Naïve Bayes, Text Mining, TF/IDF
Citation
Perpustakaan