Detection of Americans’ Behavior toward Islam on Facebook

Dublin Core

Title

Detection of Americans’ Behavior toward Islam on Facebook

Subject

analysis; behavior; document frequency; frequency-inverse; logistic regession; sentiment analysis.

Description

Social network websites have become a rich place for detecting and analyzing people’s attitudes, perceptions, and feelings towards news, products, and other real-world issues. Facebook is a popular platform among different age groups and countries and is generally used to convey ideas about certain topics based on likes, comments and sharing. In recent years, one of the most controversial topics were the idea behind Islamophobia and other ideas built in people’s minds about Islam around the world. This research studied the public opinion of American citizens about Islam during the presidency of Donald Trump, as that period was rich in diversity of opinion between his supporters and detractors. In this paper, sentiment analysis was used to analyze American citizens’ behavior towards posts about Islam during Trump’s presidency in various states across the United States. Sentiment analysis was performed on Facebook posts and comments extracted from American news channels from the year 2017. Several machine learning methods were used to detect the polarity in the dataset. The highest classification accuracy among the classifiers used in this research was achieved using a logistic regression classifier, reaching 84%.

Creator

Qusai Abuein1, Mohammed Q. Shatnawi1,2,* & Lujain Ghazalat

Source

https://journals.itb.ac.id/index.php/jictra/article/view/18292/6081

Publisher

Jordan University of Science and Technology, Jordan

Date

27 Desember 2022

Contributor

Fajar bagus W

Format

PDF

Language

English

Type

Text

Files

Citation

Qusai Abuein1, Mohammed Q. Shatnawi1,2,* & Lujain Ghazalat, “Detection of Americans’ Behavior toward Islam on Facebook,” Repository Horizon University Indonesia, accessed March 15, 2025, https://repository.horizon.ac.id/items/show/7027.