ANALISIS SENTIMEN TWITTER DENGAN MENGGUNAKAN ALGORITMA NAÏVE BAYES DENGAN SELEKSI FITUR ALGORITMA GENETIKA (Studi Kasus: Pembelajaran Online Pada Masa Pandemi)

AHMAD MU’ALLAL HIFNI, NIM. 17106050034 (2021) ANALISIS SENTIMEN TWITTER DENGAN MENGGUNAKAN ALGORITMA NAÏVE BAYES DENGAN SELEKSI FITUR ALGORITMA GENETIKA (Studi Kasus: Pembelajaran Online Pada Masa Pandemi). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS SENTIMEN TWITTER DENGAN MENGGUNAKAN ALGORITMA NAÏVE BAYES DENGAN SELEKSI FITUR ALGORITMA GENETIKA (Studi Kasus: Pembelajaran Online Pada Masa Pandemi)) 17106050034_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (ANALISIS SENTIMEN TWITTER DENGAN MENGGUNAKAN ALGORITMA NAÏVE BAYES DENGAN SELEKSI FITUR ALGORITMA GENETIKA (Studi Kasus: Pembelajaran Online Pada Masa Pandemi)) 17106050034_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (4MB)

Abstract

The learning system has changed since the Covid-19 pandemic. The learning system which is normally carried out face-to-face has turned into online or online learning. Students have many opinions and responses to online learning during this pandemic, especially those written on Twitter social media. This student opinion has the potential to be researched in the development of online learning. Unfortunately, this tweets data is unstructured so it needs to be processed first. This research uses the Naïve Bayes Algorithm that uses the feature selection of the Genetic Algorithm. The algorithm is tested by processing tweets that have gone through the preprocessing stage, then the words in the tweets are given weights using TF-IDF, then the analysis process is carried out with both algorithms. The results of research conducted using the Naïve Bayes Algorithm with feature selection of the Genetic Algorithm with a population value of 250 and a generation value of 25 obtained an accuracy value of 67%, while for the Naive Bayes Algorithm without using feature selection, an accuracy value of 76% was obtained. And students who have a positive opinion of 32.2%, students with a neutral opinion of 14.7%, and students with a negative opinion of 53% for the Naïve Bayes algorithm without feature selection.

Share this knowledge with your friends :

Actions (login required)

View Item