Sentiment Analysis of Learning from Home During Pandemic Covid-19 in Indonesia

DOI

10.2991/assehr.k.210805.073How to use a DOI?

Keywords

Twitter, sentiment analysis, Covid-19, learning from home, machine learning, classification

Abstract

Covid-19 pandemic in Indonesia impacting education sector issuance policy for learning from home. This policy has many responses from the public on social media like Twitter. There are many types of methods using machine learning approach to classify and predict sentiment analysis about this topic on Twitter. A good method with the best performance is required to increase chances of getting correct prediction and classification. To answer these problems and solve these needs, this research aims to find the best performing classification and prediction methods for analysis. This research objective is also expected to provide an overview for practitioners as a reference for researchers in conducting research. This research used a dataset consisting of 71.70% negative sentiment, 11.78% positive sentiment, and 16.52% neutral from 27708 response data on Twitter. Sentiment grouping by 3 annotators has an average annotation agreement level of 80.57%. Logistic regression produced the best performance compared to 6 other methods with performance accuracy 98.03% and f1-score 92.69%. Information extraction that causes negative sentiment mostly comes from the process of self-adaptation to the learning from home policy. Meanwhile, the provision of the Ministry of Education and Culture’s quota assistance has a major influence on positive sentiment. Information extraction that affects to this sentiment can be used by the government to improve the application of learning from home policies such as providing internet quotas which have been proven to affect positive sentiment.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Bambang Widoyono
AU  - Indra Budi
AU  - Prabu Kresna Putra
AU  - Aris Budi Santoso
PY  - 2021
DA  - 2021/08/08
TI  - Sentiment Analysis of Learning from Home During Pandemic Covid-19 in Indonesia
BT  - Proceedings of the International Conference on Economics, Business, Social, and Humanities (ICEBSH 2021)
PB  - Atlantis Press
SP  - 464
EP  - 472
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.210805.073
DO  - 10.2991/assehr.k.210805.073
ID  - Widoyono2021
ER  -