Visual Themes and Sentiment on Social Networks To Aid First Responders During Crisis Events

• DOI: 10.48550/arxiv.1610.07772
• Publication Date: 2016-01-01
• Authors: Prateek Dewan; Varun Bharadhwaj; Aditi Mithal; Anshuman Suri; Ponnurangam Kumar...
• Keywords: Sentiment Analysis; Misinformation; Content (measure theory)

Online Social Networks explode with activity whenever a crisis event takes place. Most content generated as part of this is mixture text and images, particularly useful for first responders to identify popular topics interest gauge the pulse sentiment citizens. While multiple researchers have used identify, analyze measure themes public during such events, little work has explored visual floating on networks in form inspired by them. Given potential influencing users' thoughts emotions, we perform large scale analysis compare across images textual posted Facebook terror attacks that took place Paris 2015. Using state-of-the-art image summarization techniques, discovered which were but not identifiable through text. We uncovered instances misinformation false flag (conspiracy) theories among themes, prominent user content, can be particular inter- est responders. Our also revealed while after reflected negative sentiment, positive sentiment. To best our knowledge, study social event.