Cross-Language Evolution of Divergent Collective Memory Around the Arab Spring

Read original: arXiv:2404.10706 - Published 4/17/2024 by H. Laurie Jones, Brian C. Keegan
Total Score

0

Cross-Language Evolution of Divergent Collective Memory Around the Arab Spring

Sign in to get full access

or

If you already have an account, we'll log you in

Overview

  • This research paper examines how collective memories around the Arab Spring protests evolved differently across language versions of Wikipedia.
  • The study tracked changes in the salience of key events related to the Arab Spring over time, comparing the English, Arabic, and French Wikipedia articles.
  • The findings reveal divergent narratives emerging in the different language communities, highlighting the role of language in shaping collective memory.

Plain English Explanation

The Arab Spring was a series of protests and uprisings that swept across the Middle East and North Africa in the early 2010s. This research looks at how the collective memories of these events have evolved differently across language versions of the online encyclopedia Wikipedia.

The researchers analyzed how the significance or "salience" of key Arab Spring events, like the Tunisian revolution and the Egyptian protests, was represented over time in the English, Arabic, and French Wikipedia articles. They found that the narratives around these events diverged across the language communities.

For example, the Tunisian revolution was more prominently featured in the French Wikipedia articles compared to the English and Arabic versions. And the Egyptian protests were emphasized more heavily in the Arabic articles. This suggests that the way people remember and understand major historical events can be shaped by the language they use to access information about them.

The study highlights how language can create different collective memories, even for the same set of events. This is an important finding, as Wikipedia is a major source of information that helps shape public knowledge and discourse around current events and history.

Technical Explanation

The researchers used several quantitative measures to track the longitudinal evolution of event salience in the different language versions of Wikipedia. These included:

  1. Size: The length of the Wikipedia articles related to key Arab Spring events.
  2. Centrality: How central or important the events were positioned within the overall structure of the Wikipedia articles.
  3. Attention: The volume of edits and revisions made to the articles over time, indicating the level of ongoing attention and discussion.

By comparing these metrics across the English, Arabic, and French Wikipedia over the course of several years, the study revealed distinct trajectories in how the collective memory of the Arab Spring unfolded in each language community.

This research builds on previous work on language-agnostic modeling of Wikipedia content quality and unveiling global narratives through multilingual Twitter datasets. It also connects to efforts to promote linguistic diversity in the digital sphere and scale up multilingual analysis of news framing.

Critical Analysis

The paper acknowledges some limitations in its approach, such as the challenges of comparing content across different language versions of Wikipedia due to differences in editing practices and community dynamics. Additionally, the study focused only on a subset of key Arab Spring events, rather than attempting to capture the full breadth of coverage.

One potential area for further research would be to expand the analysis to include other language editions of Wikipedia, as well as to explore how these divergent collective memories are reflected in other multilingual information sources, such as political knowledge graphs. This could provide a more comprehensive understanding of how language shapes the global discourse around major historical events.

Conclusion

This study offers valuable insights into how language influences the collective memory of significant political and social upheavals like the Arab Spring. By tracing the evolution of event salience across different Wikipedia language editions, the researchers have demonstrated how narratives can diverge even when the underlying events are the same.

These findings have important implications for our understanding of how knowledge and historical perspectives are constructed and shared in the digital age. As Wikipedia becomes an increasingly influential source of information, it is crucial to recognize the role that language, and the communities it shapes, plays in shaping collective memory and public discourse.



This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!

Follow @aimodelsfyi on 𝕏 →

Related Papers

Cross-Language Evolution of Divergent Collective Memory Around the Arab Spring
Total Score

0

Cross-Language Evolution of Divergent Collective Memory Around the Arab Spring

H. Laurie Jones, Brian C. Keegan

The Arab Spring was a historic set of protests beginning in 2011 that toppled governments and led to major conflicts. Collective memories of events like these can vary significantly across social contexts in response to political, cultural, and linguistic factors. While Wikipedia plays an important role in documenting both historic and current events, little attention has been given to how Wikipedia articles, created in the aftermath of major events, continue to evolve over years or decades. Using the archived content of Arab Spring-related topics across the Arabic and English Wikipedias between 2011 and 2024, we define and evaluate multilingual measures of event salience, deliberation, contextualization, and consolidation of collective memory surrounding the Arab Spring. Our findings about the temporal evolution of the Wikipedia articles' content similarity across languages has implications for theorizing about online collective memory processes and evaluating linguistic models trained on these data.

Read more

4/17/2024

ArMeme: Propagandistic Content in Arabic Memes
Total Score

0

ArMeme: Propagandistic Content in Arabic Memes

Firoj Alam, Abul Hasnat, Fatema Ahmed, Md Arid Hasan, Maram Hasanain

With the rise of digital communication, memes have become a significant medium for cultural and political expression that is often used to mislead audiences. Identification of such misleading and persuasive multimodal content has become more important among various stakeholders, including social media platforms, policymakers, and the broader society as they often cause harm to individuals, organizations, and/or society. While there has been effort to develop AI-based automatic systems for resource-rich languages (e.g., English), it is relatively little to none for medium to low resource languages. In this study, we focused on developing an Arabic memes dataset with manual annotations of propagandistic content. We annotated ~6K Arabic memes collected from various social media platforms, which is a first resource for Arabic multimodal research. We provide a comprehensive analysis aiming to develop computational tools for their detection. We will make them publicly available for the community.

Read more

6/7/2024

🎲

Total Score

0

Sina at FigNews 2024: Multilingual Datasets Annotated with Bias and Propaganda

Lina Duaibes, Areej Jaber, Mustafa Jarrar, Ahmad Qadi, Mais Qandeel

The proliferation of bias and propaganda on social media is an increasingly significant concern, leading to the development of techniques for automatic detection. This article presents a multilingual corpus of 12, 000 Facebook posts fully annotated for bias and propaganda. The corpus was created as part of the FigNews 2024 Shared Task on News Media Narratives for framing the Israeli War on Gaza. It covers various events during the War from October 7, 2023 to January 31, 2024. The corpus comprises 12, 000 posts in five languages (Arabic, Hebrew, English, French, and Hindi), with 2, 400 posts for each language. The annotation process involved 10 graduate students specializing in Law. The Inter-Annotator Agreement (IAA) was used to evaluate the annotations of the corpus, with an average IAA of 80.8% for bias and 70.15% for propaganda annotations. Our team was ranked among the bestperforming teams in both Bias and Propaganda subtasks. The corpus is open-source and available at https://sina.birzeit.edu/fada

Read more

7/15/2024

🌐

Total Score

0

Highly engaging events reveal semantic and temporal compression in online community discourse

Antonio Desiderio, Anna Mancini, Giulio Cimini, Riccardo Di Clemente

People nowadays express their opinions in online spaces, using different forms of interactions such as posting, sharing and discussing with one another. How do these digital traces change in response to events happening in the real world? We leverage Reddit conversation data, exploiting its community-based structure, to elucidate how offline events influence online user interactions and behavior. Online conversations, as posts and comments, are analysed along their temporal and semantic dimensions. Conversations tend to become repetitive with a more limited vocabulary, develop at a faster pace and feature heightened emotions. As the event approaches, the shifts occurring in conversations are reflected in the users' dynamics. Users become more active and they exchange information with a growing audience, despite using a less rich vocabulary and repetitive messages. The recurring patterns we discovered are persistent across a wide range of events and several contexts, representing a fingerprint of how online dynamics change in response to real-world occurrences.

Read more

7/24/2024