Persepsi Publik Tentang Pembelajaran Daring di Indonesia: Studi Menggunakan ELK Stack untuk Analisis Sentimen di Twitter

Oktavianto, Andri (2020) Persepsi Publik Tentang Pembelajaran Daring di Indonesia: Studi Menggunakan ELK Stack untuk Analisis Sentimen di Twitter. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 09111640000060-Undergraduate_Thesis.pdf]
Preview
Text
09111640000060-Undergraduate_Thesis.pdf

Download (1MB) | Preview

Abstract

Penelitian ini menyelidiki persepsi publik tentang aplikasi pembelajaran daring di Indonesia. Banyak studi tentang pembelajaran daring dilakukan di negara maju dan hanya sedikit di negara berkembang. Selain itu, penelitian-penelitian tersebut menggunakan pendekatan kualitatif yang berarti hasil penelitian hanya bisa diterapkan di latar yang spesifik. Sementara penelitian tradisional menggunakan survei untuk memahami persepsi orang terhadap suatu topik membutuhkan banyak waktu dan upaya, penelitian ini menggunakan cara yang efisien dan efektif untuk mengumpulkan pendapat dan kemudian menganalisis sentimennya menggunakan Elasticsearch, Logstash (ELK stack) dan bahasa pemrograman Python. Algoritma Naïve Bayes digunakan untuk analisis sentimen dan ELK stack untuk mengumpulkan dan menyimpan tweets dari Twitter. Dengan ELK stack, penelitian ini berhasil mengumpulkan 133.477 tweets yang terkait dengan pembelajaran daring. Dari hasil analisis sentimen ditemukan sebanyak 98.3% tweets memiliki sentimen positif terhadap pembelajaran daring, hal ini menjadi bukti bahwa pelajar di Indonesia memiliki persepsi positif terhadap pembelajaran daring. Penelitian ini juga menghasilkan beberapa wawasan mengenai preferensi aplikasi pembelajaran daring pelajar di Indonesia yang berguna bagi penyedia layanan pembelajaran daring. Penelitian ini sebelumnya dipublikasikan di International Journal of Emerging Technologies in Learning (iJET) tahun 2020.

============================================================================

This study investigates people's perceptions about online learning applications in Indonesia. Previous studies on online learning was conducted in developed countries and only a little in developing countries. In addition, these studies use a qualitative approach which means that the research results can only be applied in a specific setting. While traditional research using surveys to understand people's perceptions towards a topic takes a lot of time and effort, this research use an efficient and effective way to gather opinions and then analyze their sentiments using Elasticsearch, Logstash, Kibana (ELK stack), and Python programming language. This research use Naïve Bayes algorithm for sentiment analysis and the ELK stack for collecting & saving tweets from Twitter. Using ELK stack, this study managed to collect 133,477 tweets related to online learning. From the results of sentiment analysis, we found that 98.3% of tweets have positive sentiments towards online learning, this is an evidence that students in Indonesia have a positive perception towards online learning. This research also generates some insights regarding student online learning application preferences in Indonesia that are useful for online learning service providers. This research was previously published in the International Journal of Emerging Technologies in Learning (iJET) in 2020.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: ELK stack, Indonesia, Online learning, Sentiment Analysis, Twitter, Analisis Sentimen, ELK stack, Indonesia, Pembelajaran Daring, Twitter.
Subjects: L Education > L Education (General)
Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions: Faculty of Creative Design and Digital Business (CREABIZ) > Business Management > 61205-(S1) Undergraduate Thesis
Depositing User: Andri Oktavianto
Date Deposited: 13 Aug 2020 06:40
Last Modified: 05 Jun 2023 14:40
URI: http://repository.its.ac.id/id/eprint/77965

Actions (login required)

View Item View Item