Detecting political biases of named entities and hashtags on Twitter

EPJ Data Science

Regular Article

Yizhou Sun¹  Yining Wang¹  Zhiping Xiao¹  Wen Hong Lam¹  Jeffrey Zhu¹  Mason A. Porter²  Pei Zhou³ 

[1] Department of Computer Science, University of California, Los Angeles, 580 Portola Plaza, 90095, Los Angeles, California, United States of America;Department of Mathematics, University California, Los Angeles, 520 Portola Plaza, 90095, Los Angeles, California, United States of America;Santa Fe Institute, 1399 Hyde Park Road, 87501, Santa Fe, New Mexico, United States of America;Information Sciences Institute, University of Southern California, Marina del Rey, 90292, Los Angeles, California, United States of America;

关键词: Political-polarity detection;    Word embeddings;    Multi-task learning;    Adversarial training;    Data sets;   

DOI  :  10.1140/epjds/s13688-023-00386-6

 received in 2022-09-19, accepted in 2023-03-31,  发布年份 2023

来源: Springer

【 摘 要 】

Ideological divisions in the United States have become increasingly prominent in daily communication. Accordingly, there has been much research on political polarization, including many recent efforts that take a computational perspective. By detecting political biases in a text document, one can attempt to discern and describe its polarity. Intuitively, the named entities (i.e., the nouns and the phrases that act as nouns) and hashtags in text often carry information about political views. For example, people who use the term “pro-choice” are likely to be liberal and people who use the term “pro-life” are likely to be conservative. In this paper, we seek to reveal political polarities in social-media text data and to quantify these polarities by explicitly assigning a polarity score to entities and hashtags. Although this idea is straightforward, it is difficult to perform such inference in a trustworthy quantitative way. Key challenges include the small number of known labels, the continuous spectrum of political views, and the preservation of both a polarity score and a polarity-neutral semantic meaning in an embedding vector of words. To attempt to overcome these challenges, we propose the Polarity-aware Embedding Multi-task learning (PEM) model. This model consists of (1) a self-supervised context-preservation task, (2) an attention-based tweet-level polarity-inference task, and (3) an adversarial learning task that promotes independence between an embedding’s polarity component and its semantic component. Our experimental results demonstrate that our PEM model can successfully learn polarity-aware embeddings that perform well at tweet-level and account-level classification tasks. We examine a variety of applications—including a study of spatial and temporal distributions of polarities and a comparison between tweets from Twitter and posts from Parler—and we thereby demonstrate the effectiveness of our PEM model. We also discuss important limitations of our work and encourage caution when applying the PEM model to real-world scenarios.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

Fig. 3

Fig. 1

Fig. 1

Fig. 1

41116_2023_37_Article_IEq14.gif

41116_2023_37_Article_IEq13.gif

Fig. 1

41116_2023_37_Article_IEq12.gif

Fig. 2

41116_2023_37_Article_IEq10.gif

41116_2023_37_Article_IEq9.gif

41116_2023_37_Article_IEq7.gif

41116_2023_37_Article_IEq6.gif

41116_2023_37_Article_IEq4.gif

41116_2023_37_Article_IEq3.gif

41116_2023_37_Article_IEq2.gif

41116_2023_37_Article_IEq1.gif

Fig. 1

Fig. 1

Fig. 6

Fig. 2

Fig. 24

Fig. 7

Fig. 2

Fig. 4

Fig. 1

Fig. 3

Fig. 1

Fig. 6

Fig. 12

Fig. 1

Fig. 4

Fig. 5

Fig. 6

Fig. 5

Scheme 1

Fig. 1

Fig. 6

Fig. 1

Fig. 22

Fig. 4

40854_2023_491_Article_IEq17.gif

Fig. 2

Fig. 21

40854_2023_491_Article_IEq12.gif

Fig. 1

Fig. 5

Fig. 20

Fig. 2

Fig. 5

40854_2023_491_Article_IEq4.gif

Fig. 3

Fig. 3

Fig. 4

Fig. 4

Fig. 4

Fig. 1

Fig. 2

Fig. 1

Fig. 11

Fig. 17

Fig. 10

Fig. 16

Fig. 3

Fig. 9

Fig. 2

12936_2023_4634_Article_IEq5.gif

12936_2023_4634_Article_IEq2.gif

Fig. 3

Fig. 1

Fig. 2

Fig. 4

Fig. 2

Fig. 1

Fig. 13

Fig. 8

Fig. 1

Fig. 1

Fig. 3

Fig. 8

Fig. 1

Fig. 12

Fig. 11

Fig. 1

Fig. 10

Fig. 9

Fig. 8

Fig. 8

Fig. 1

Fig. 7

Fig. 7

12888_2023_4863_Article_IEq1.gif

Fig. 3

Fig. 5

Fig. 4

Fig. 3

Fig. 2

Fig. 6

Fig. 5

13690_2023_1130_Article_IEq62.gif

40517_2023_252_Article_IEq8.gif