Scholarlink

The Russo-Ukrainian conflict underscores challenges in obtaining reliable firsthand accounts. Traditional methods such as satellite imagery and journalism fall short due to limited access to zones. Secure social media platforms such as Telegram offer safer communication from conflict zones but lack effective message grouping, hindering insight collection. The proposed framework aims to enhance firsthand account gathering by crowdsourcing secure social media data. We gathered 250,000 Telegram messages on the conflict and developed a language model-based framework to identify contextual groupings. Evaluation reveals 477 new groupings from 13 news sources, enriching firsthand information. This research emphasizes the significance of secure social media crowdsourcing in conflict zones, paving the way for future advancements.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Leveraging Secure Social Media Crowdsourcing for Gathering Firsthand Account in Conflict Zones

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Abanisenioluwa Orojo,
  • Pranish Bhagat,
  • John Wilburn,
  • Michael Donahoo,
  • Nishant Vishwamitra

摘要

The Russo-Ukrainian conflict underscores challenges in obtaining reliable firsthand accounts. Traditional methods such as satellite imagery and journalism fall short due to limited access to zones. Secure social media platforms such as Telegram offer safer communication from conflict zones but lack effective message grouping, hindering insight collection. The proposed framework aims to enhance firsthand account gathering by crowdsourcing secure social media data. We gathered 250,000 Telegram messages on the conflict and developed a language model-based framework to identify contextual groupings. Evaluation reveals 477 new groupings from 13 news sources, enriching firsthand information. This research emphasizes the significance of secure social media crowdsourcing in conflict zones, paving the way for future advancements.