Presence of Anti-Uyghur Influence Operations in Xinjiang
Session Number
Project ID: CMPS 10
Advisor(s)
Dr. Courtland VanDam; Massachusetts Institute of Technology, Lincoln Laboratory
Discipline
Computer Science
Start Date
19-4-2023 10:50 AM
End Date
19-4-2023 11:05 AM
Abstract
The purpose of this study is to analyze patterns of Twitter to determine the existence of influence operations and their relation to the current narrative of Uyghur oppression in Xinjiang. Our analysis focuses on the Twitter Moderation Research Consortium (TMRC) of flagged accounts marked as spam, or “persistent platform manipulation campaigns,” consisting of over 2,048 accounts and 31,269 tweets. We analyze influence operations using natural language processing libraries, specifically sharp changes in topics, frequency of tweets, and sentiment of tweets. Namely, we generated word clouds, Latent Dirichlet Allocation topic models, and sentiment analysis graphs to answer our research questions. We found a targeted impact against Uyghur Muslims in Xinjiang as tweet sentiment from the region sharply drops at keystone anti-Xinjiang and anti-Chinese events in the media as topics shifted to Uyghur oppression in the region, indicating a coordinated attempt at shifting the controversial narrative and the sentiment surrounding it.
Presence of Anti-Uyghur Influence Operations in Xinjiang
The purpose of this study is to analyze patterns of Twitter to determine the existence of influence operations and their relation to the current narrative of Uyghur oppression in Xinjiang. Our analysis focuses on the Twitter Moderation Research Consortium (TMRC) of flagged accounts marked as spam, or “persistent platform manipulation campaigns,” consisting of over 2,048 accounts and 31,269 tweets. We analyze influence operations using natural language processing libraries, specifically sharp changes in topics, frequency of tweets, and sentiment of tweets. Namely, we generated word clouds, Latent Dirichlet Allocation topic models, and sentiment analysis graphs to answer our research questions. We found a targeted impact against Uyghur Muslims in Xinjiang as tweet sentiment from the region sharply drops at keystone anti-Xinjiang and anti-Chinese events in the media as topics shifted to Uyghur oppression in the region, indicating a coordinated attempt at shifting the controversial narrative and the sentiment surrounding it.
