Data Briefs
Lebanon Data Signaled Escalation Before Ceasefire
Signals of escalation appeared in Lebanon-related posts on X before the ceasefire announcement, preceding developments later reported on the ground.
Mohamed Soufan
Computational Researcher and Software Engineer analyzing large-scale social media data to understand online discourse. Uses data science and computational methods to uncover patterns in digital communication and public debate.
Data Briefs
Lebanon Strike Reports 5× Higher Before Ceasefire
Strike reports in Lebanon surged sharply before the April 8 ceasefire announcement, peaking prior to official confirmation and remaining elevated during it.
Analysis
Lebanon Strike Mentions Jump 5× Pre-Ceasefire
Analysis of 5,142 posts on X shows strike reports surged before the ceasefire announcement and remained elevated during it.
Data Briefs
Top 1% Dominate Hezbollah Discourse on X, Study Finds
A small minority of users capture most engagement in Hezbollah-related discussions on X, revealing a highly concentrated attention structure.
Analysis
Inside the Hidden Structure of Political Discourse on X
A study of Arabic-language political discussions on X shows that while thousands participate, attention is highly concentrated—just 1% of users drive over 60% of engagement.
Data Briefs
Ain Saadeh Incident: X Peaked Before Media Coverage
The Ain Saadeh timeline shows how quickly online discussion developed compared to media coverage.
Data Briefs
Ain Saadeh: Political Posts Got 4.6× More Engagement
Political posts represented only a small share of posts about the Ain Saadeh incident on X, but they received far higher engagement than other types of content.
Analysis
Ain Saadeh on X: Politics Drove Engagement, Not News
After the Ain Saadeh incident, strike claims dominated what people posted on X, but political content dominated what people paid attention to.
Papers
Attention Concentration in Online Political Discussions on X
This study analyzes engagement in online political discussions on X and finds that attention is highly concentrated, with a small number of users receiving most engagement.
Data Briefs
Uncertain Tweets Get More Likes, Reposts, and Replies
Tweets expressing uncertainty receive higher engagement across likes, reposts, and replies in Arabic-language discussions about Lebanon on X.
Data Briefs
30% of Arabic Tweets About Lebanon Contain Uncertainty
Nearly one-third of Arabic-language tweets about Lebanon contain linguistic uncertainty, showing that tentative language is a common feature of online discussions.
Data Briefs
Uncertain Tweets Get 25% More Engagement on X
Analysis of Arabic-language tweets shows that tweets expressing uncertainty receive higher engagement, particularly in replies, on X.