Home // HUSO 2023, The Ninth International Conference on Human and Social Analytics // View article

Leveraging Topic Modeling and Toxicity Analysis to Understand China-Uyghur Conflicts

Authors:
Connice Trimmingham
Samuel Dayo Banjo
Nitin Agarwal

Keywords: Information operations; Indo-Pacific; Twitter; Social Media; Uyghur; Toxicity Analysis.

Abstract:
The growth of social networking sites, coupled with the widespread use of mobile technology, has led to the spread of various forms of toxicity. Although social media platforms provide valuable tools for meaningful interactions, political arguments, often fraught with complex mix of emotions, can quickly devolve into flame wars or partisan bickering. This article shifts attention eastward to examine how the media/information environment is being manipulated for advancing political agendas in the Indo-Pacific region. We analyzed 3,239,249 tweets discussing issues related to China and Uyghur. We explained the user relation phenomena by assessing their emerging social structures. We extracted influential topics using the Latent Dirichlet Allocation (LDA) topic modeling approach. Toxicity analysis and bot assessment were performed to examine the nature of discourse about the China and Uyghur issues. Our findings indicate a strong correlation between tweets with high toxicity and bot activity, particularly in relation to emerging events such as the existence of internment camps and news about forced Uyghur laborers in China and the Chinese Communist Party network.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2023

Publication date: March 13, 2023

Published in: conference

ISSN: 2519-8351

ISBN: 978-1-68558-066-7

Location: Barcelona, Spain

Dates: from March 13, 2023 to March 17, 2023